OECD Employment Outlook 2018

Report

OECD Employment Outlook

4 July 2018

Available in:

English
français

Download PDF

Chapter 4. Back to work: Lessons from nine country case studies of policies to assist displaced workers

Abstract

This chapter analyses how best labour market programmes can reduce the costs borne by workers who lose their jobs due to business closings or other economic reasons (“displaced workers”). The chapter shows that a considerable number of workers are displaced every year and that many in this group – especially older workers in blue‑collar jobs – experience large earnings losses due to both long periods out of work and re‑employment at a lower wage. The chapter draws upon detailed case studies of policies to assist displaced workers in nine OECD countries and provides many examples of the effective use of active labour market policies and unemployment benefits to ensure that the labour market adjustment costs inherent to a dynamic economy are kept as low as possible and that these costs are not unfairly concentrated on the displaced workers who have the most limited job mobility.

The statistical data for Israel are supplied by and under the responsibility of the relevant Israeli authorities. The use of such data by the OECD is without prejudice to the status of the Golan Heights, East Jerusalem and Israeli settlements in the West Bank under the terms of international law.

Key findings

A dynamic economy requires a fluid labour market in which workers are continually moving from shrinking to growing firms. Indeed, labour reallocation is an integral part of the “creative destruction” process that underlies economic growth and rising living standards. A considerable number of workers who lose their jobs to economic change (“displaced workersˮ) nevertheless experience significant income losses and other hardships, and these costs need to be kept as low as possible if the net benefits of growth are to be maximised and shared in an equitable manner. One way that governments can reduce the costs of labour market restructuring is by improving the re‑employment assistance and income support that labour market programmes provide to displaced workers. This chapter discusses how best that can be achieved, highlighting the general policy lessons from the OECD’s Back to Work reviews of nine countries which analysed policies to improve the re‑employment prospects of displaced workers. The chapter first examines the scale of job displacement and the labour market problems encountered by those affected:

Job displacement – defined as a permanent economic dismissal affecting a worker who has at least one year of job tenure – is quite common. Between 1% and 7% of the workforce is displaced annually, implying a significant probability that a typical worker will experience displacement one or more times during her working life. Nonetheless, job displacement accounts for only a modest share of all job separations, many of which are voluntarily initiated by the worker.
A considerable number of displaced workers find a suitable new job rapidly, but the majority experience significant losses of income and potentially would benefit from re‑employment assistance and income support. Income losses are particularly large during the period of joblessness that immediately follows displacement in the majority of cases. In certain countries earnings can fall by up to 50% on the year of dismissal and remain up to 10% below pre‑displacement years even four years after being laid‑off. However, income losses can continue after displaced workers are re‑employed, because wages in post‑displacement jobs are often lower than those from the lost jobs. The risks of long‑term joblessness and large earnings losses after re‑employment are particularly significant for older and long‑tenure workers in blue‑collar jobs.
Displaced workers find new jobs much more rapidly in some countries than in others. Whereas nearly 90% are re‑employed within a year in Finland and Sweden, the corresponding figure for France and Portugal is about 30%. The earnings losses of displaced workers also vary considerably across the countries analysed. These differences suggest that national labour market policies and institutions can have a significant impact on the adjustment costs borne by displaced workers.
The costs faced by displaced workers are also highly variable within the same country. The re‑employment and retraining support offered to each worker therefore needs to be calibrated to reflect individual needs. However, these needs may be difficult to assess in a timely manner in practice.
There is a surprising scarcity of rigorous evaluation evidence concerning what works for displaced workers. Additional evaluations of the effectiveness of the different policy options that are discussed here would be very helpful.

When feasible, proactive actions can play some useful role in limiting the cost of job displacement. These can take the form of:

Preventive measures. There is a potential, albeit limited, role for measures to prevent layoffs that are not socially efficient, either by effectively taxing layoffs (e.g. through higher unemployment insurance contributions for employers who lay off workers or employment protection legislation rules that impose costs on firms dismissing permanent workers) or by using a short‑time work scheme to encourage employers facing temporary difficulties to preserve jobs that are viable in the long run. However, a light touch with preventive measures is required, especially as regards employment protection legislation, so as to preserve efficiency in the reallocation process and avoid discouraging job creation.
Early intervention measures. A crucial difference between displaced workers and most other groups served by the public employment service (PES) is that it is often possible to initiate re‑employment services during the notice period prior to displacement. Rapid response services, such as setting up a temporary PES office in a factory that will soon close, facilitate the timely delivery of re‑employment services which can jump‑start the adjustment process. These early intervention services can be quite effective. However, they are not used as widely as would be desirable, often being limited to workers affected by mass layoffs. In Sweden, the job security councils that are operated by the social partners demonstrate the feasibility of offering effective early intervention measures to all displaced workers, including those affected by individual or small‑scale layoffs, when employers and unions are constructively engaged. It is important to require that employers provide at least a minimally adequate period of advance notice of layoffs whenever possible, while ensuring that notified workers be the focus of outreach initiatives by the PES or be required to register with the PES as soon as they are notified.

An effective national activation strategy to get people into work provides a solid foundation for promoting rapid the re‑employment of displaced workers while a well‑designed unemployment benefits scheme is key to providing adequate income security. However, policy also needs to take into account of the specific barriers to re‑employment that often confront displaced workers – such as obsolete skills and the absence of recent job search experience, while also leveraging their advantages – notably a history of stable employment and strong labour force attachment. Adapting re‑employment and income supports for this group raises issues in a number of areas:

Active labour market programmes (ALMPs). While all displaced workers should benefit from prompt access to basic job search services, some will require more intensive re‑employment services or retraining. One key challenge is to identify this smaller group rapidly and offer them intensive services when these are most effective, rather than only after a long period of unemployment as is frequently the case. Two additional challenges are to reduce the often large inequities in the access of different groups of displaced workers to re‑employment help and to rapidly scale‑up re‑employment services when there is an upsurge in displacement, either locally or nationally. Re‑employment services for displaced workers are typically delivered by a combination of general ALMPs and programmes targeted at this specific group and a pragmatic mix of general and targeted programmes is ideal. In countries where many displaced workers can only access unemployment benefits after a long delay (or never), it is important to decouple the initiation of contact with the PES and access to re‑employment support from the timing of benefit receipt.
Income support. In most countries with unemployment insurance systems, the benefit entitlements of displaced workers tend to be relatively high because they have more stable employment histories than most other unemployed persons. Nonetheless, benefits only provide compensation for a small proportion of earnings losses, especially for workers who experience a long spell of unemployment or become re‑employed at a significantly lower wage or in a part‑time job. Providing adequate income support for displaced workers, while also encouraging rapid re‑employment, requires good programme design. For example, a temporary wage supplement can be offered to displaced workers who return to work rapidly by accepting a new job at a lower pay level. Much remains to be learned, however, about how best workers can be insured against the earnings losses due to displacement while minimising adverse impacts on job search incentives.

A broader set of policies can contribute to the successful management of labour market restructuring, including by helping to lower the costs of job displacement, although these lie beyond the scope of this chapter. Important examples include policies to foster economic revitalisation in regions that have been hit hard by mass layoffs and an effective national system to anticipate and meet changing job skill requirements.

Introduction

OECD labour markets are characterised by their dynamism. Each year, more than 20% of jobs, on average, are created and/or destroyed, and around one‑third of all workers are hired by and/or separate from their employer (OECD, 2009[1]). These large job and worker flows are a reflection of a continuous process of labour reallocation that contributes to productivity gains and rising living standards. However, high job turnover is also a source of insecurity for workers, especially those who lose their jobs because their employer downsizes its workforce or goes out of business altogether, a group that labour market researchers typically label as “displaced workersˮ.1 An important challenge facing OECD governments is to nurture labour market dynamism while keeping the adjustment costs borne by displaced workers as low as possible. This chapter analyses how labour market programmes can contribute to meeting that policy challenge by improving the re‑employment prospects of displaced workers and compensating them for part of their lost earnings.

Assisting displaced workers has long been recognised as being an important policy goal for active and passive labour market programmes. In fact, unemployment insurance (UI) and the active labour market programmes (ALMPs) that are offered by the public employment service (or its private subcontractors), such as job placement and retraining, were developed in large part to assist workers who were laid‑off by their employers in response to changing economic conditions. The vast research literature analysing the design and operation of such programmes is highly relevant for assessing how they can assist this group in a cost‑effective manner. However, this research rarely assesses the specific situation of displaced workers, as distinct from that of other participants of labour market programmes, making it difficult to draw firm conclusions about which general measures work best for this particular group and whether they also require special support measures. This limitation is particularly unfortunate at a time of heightened public concerns about the uneven distribution of the benefits and costs associated with rapid economic change. Workers who are displaced due to globalisation, technological and other structural change figure prominently among the potential victims of economic restructuring, making it timely to assess how well labour programmes are meeting the re‑employment and income support needs of this group.2

This chapter documents the labour market services currently offered to displaced workers in OECD countries and assesses how those services could be improved. Policy reforms that reduce the costs associated with job displacement would bring significant overall efficiency and equity benefits, for example by helping workers whose jobs are automated to move quickly into other jobs that make productive use of their skills, while also helping to maintain support for open, innovative and flexible economies.3

This chapter provides an overview of the main policy lessons that emerged from a multi‑year study of job displacement, and particularly nine country reviews of policies to assist workers displaced by economic change, which the OECD conducted between 2013 and 2017.4 These Back to Work reviews provide new insights into effective policy approaches in different national contexts, while also highlighting several pitfalls to avoid. Since this chapter is intended to provide policy lessons that are general enough to apply to countries that did not participate in the nine reviews, it focuses on issues that arose in the provision of cost‑effective support to displaced workers in most or all of the countries reviewed.5 The chapter also draws upon relevant experience in other OECD countries and updates key parts of the cross‑country statistical analysis of the incidence and consequences of job displacement that was originally presented in Chapter 4 of the 2013 issue of this publication (OECD, 2013[2]) .

The remainder of the chapter is organised as follows. Section 4.1 briefly documents the incidence and consequences of job displacement and situates the labour market policies that are analysed in detail in this chapter within the broader range of policies affecting labour market adjustment costs. It also discusses the disappointing paucity of rigorous evaluation evidence concerning the cost‑effectiveness of labour market programmes for displaced workers. Section 4.2 then analyses policies that either prevent some displacements or allow the adjustment process to get underway during the notice period. Section 4.3 looks at ALMPs that are intended to assist displaced workers back into suitable jobs, including job‑search assistance, counselling and retraining. In Section 4.4, attention turns to unemployment benefits and public and private measures that compensate displaced workers for at least a part of their lost earnings. A concluding section briefly recaps the main lessons for policy while also highlighting issues that require further study.

4.1. The policy challenge

4.1.1. Overview of job displacement and the resulting costs for workers

Much of the public concern about job displacement centres on mass redundancies, such as when a factory closes or a large firm declares bankruptcy and lays off its entire workforce. Many case studies of particular mass layoffs have documented the disruption and economic hardship that result for many of the workers losing their jobs (especially older, long‑tenure workers), their families and often also for the broader community where they live, due to spill‑over effects on the local economy.6 A growing number of econometric studies have confirmed that the costs in higher unemployment and inactivity, lower wages and poorer health can be large and persistent for workers in local labour markets that experience a sudden increase in import competition, particularly blue collar workers who are initially employed in industries facing that competition (Autor, Dorn and Hanson, 2016[3]). The evidence from studies of specific mass layoffs and the local impacts of “the China shock” makes it clear that the cost of plant closings and mass layoffs can be very high. However, more comprehensive types of evidence are also required to assess the overall importance of job displacement as a source of insecurity for workers and the implications for labour market policy.

The OECD has collected harmonised statistics on job displacement in 13 OECD countries which help to characterise the overall policy challenge related to assisting displaced workers. These statistics were presented in detail in OECD (2013[2]). A partially updated version of these data are now used to provide a brief overview of the incidence and consequences of job displacement which helps define the policy challenge confronting labour market programmes intended to assist displaced workers.7

How large is the risk of displacement and who is most affected?

Figure 4.1 shows that the annual incidence of job displacement has ranged from 1% to 7% of dependent (wage and salary) employment since 2003 in the 13 OECD countries for which harmonised data could be assembled. Unfortunately, the considerable cross‑country differences in the estimated displacement rates are difficult to interpret because they reflect a combination of real differences in the incidence of displacement (e.g. due to differences in industry structure or employment protection rules) and measurement issues related to differences in the underlying data sources and definitions.8 In order to gauge the cyclicality of the displacement rate, the average risk of displacement is shown separately for three periods: the years preceding the global financial crisis, the peak crisis years and the early post‑crisis recovery period. Not surprisingly, the displacement rate is counter‑cyclical, being highest in almost all countries during the 2009‑10 crisis period. Nonetheless, the displacement rate during the pre‑ and post‑crisis periods was approximately two‑thirds that during the crisis, indicating that the majority of job displacements reflect on‑going structural change and the changing competitive position of different firms, rather than business cycle downturns.9

These displacement rates suggest that a typical worker faces a substantial risk of experiencing at least one economic layoff over the course of her working life, helping to explain the political salience of concerns about job displacement.10 Nonetheless, total job separations considerably exceed the number of workers who are displaced, since many workers voluntarily quit their jobs each year and it is also quite common for low‑tenure workers on temporary contracts to leave the firm when their contract expires.11 Figure 4.2 shows that total separation rates in OECD countries have ranged from between 9% and 32% in recent years (Panel A), with displacements estimated to account for between 9% and 36% of total separations, in the smaller number of countries where both rates could be estimated (Panel B).12 In sum, job displacement is not a rare phenomenon and it is understandable that the threat of displacement looms large in the minds of many workers, but it also should be borne in mind that these layoffs occur within the context of high labour turnover which sees approximately one‑third of all workers being hired and/or separated from their employer each year. Nonetheless, since mid‑career job mobility is common and often voluntary, it cannot be assumed that job displacement necessarily results in large and enduring costs for all workers who experience it.

Figure 4.1. Between 1% and 7% of workers lose their jobs to economic change every year

Job displacement affects all types of workers, but some demographic groups face a greater risk of involuntary job loss than others.13 In most of the countries studied, displacement rates are higher for both workers in their 20s and older workers (aged 55‑64 years) than for prime‑age workers. Workers with less than a secondary education and those whose jobs require largely manual skills also have above‑average rates of displacement. Men are displaced more often than women in most countries, but this difference appears to reflect gender differences in the types of jobs held – notably the over‑representation of men in manufacturing, construction and manual occupations rather than any discrimination against men when it comes to dismissal. Two of the strongest patterns affecting the incidence of displacement is that lower tenure workers (1‑4 years of job tenure) and employees in smaller firms (10‑49 workers) face a significantly elevated risk of involuntary job loss.

Figure 4.2. Job displacement is only one (particularly disruptive) form of worker turnover

The finding that job displacement disproportionately affects younger and low‑tenure tenure workers, as well as workers employed by smaller firms, suggests that the experience of many displaced workers may not conform to the worst case scenario where one or more large employers close creating a situation of persistently high unemployment in the local labour market, large negative spill‑over effects on the local economy and high adjustment costs for the affected workers, many of whom are long‑tenure workers in declining occupations and sectors. Indeed, the limited evidence available from other data sources suggests that the majority of displacements are associated with smaller layoff events that are unlikely to be associated with large negative externalities depressing the local labour market, but may nonetheless imply significant adjustment costs for the affected workers:

The European Restructuring Monitor (ERM) Database provides information about large restructuring events in European countries since 2002, as compiled by a network of specialists who monitor various news sources and company announcements. The number of layoffs captured by this database during 2000‑08 can be compared with the number of displacements during the same period as captured in the OECD data underlying Figure 4.1, albeit only for seven European countries. In most countries, the mass dismissals captured in the ERM dataset represent less than 15% of all displacements. Similarly, administrative data on mass layoffs that were collected by the US Bureau of Labour Statistics until recently captured only about one‑fifth as many layoffs as the Displaced Worker Survey which collected information on all economic dismissals (OECD, 2005[6]). These comparisons suggest that a large majority of all job displacement take the form of small to medium scale layoffs, rather than mass layoffs.
Two household surveys that are designed to provide comprehensive estimates of job displacement also provide information about whether each displacement occurred as a result of a business or plant closing. The Displaced Worker Survey indicates that 37% of all displacements in the United States represented total closings during 2013‑15 (BLS, 2016[7]), while the SHARELIFE data indicate that 48% of the displacements in 13 EU countries represented total closings during 1986‑2008 (Andrews and Saia, 2017[8]). While these estimates of the shutdown share of total displacements exceed the estimates of the mass layoff share cited above, it should be noted that many business closings likely involve small businesses and thus relatively few workers.
Even if the majority of displaced workers do not fit the worst case scenario, the minority who have their working lives upended in such a manner are almost certainly a sufficiently large group to be of considerable policy concern. It should also be noted that some workers affected by individual or small scale displacements may have worked for local subcontractors of a large firm that closed and, hence, are likely to face particularly difficulties in finding a suitable new job where they live.

The consequences of displacement for the affected workers

While a considerable number of workers are displaced each year, the implications for labour market programmes depends crucially on the consequences for the affected workers; in particular, whether most workers experiencing a displacement are able to find new jobs quickly that are comparable to the lost jobs or, instead, experience long periods out of work or significant earnings losses even once re‑employed.

Figure 4.3 presents re‑employment rates within 1 and 2 years of experiencing an economic dismissal.14 While some displaced workers return to work relatively rapidly, many remain jobless for an extended period. Dramatic differences in the speed of re‑employment are also apparent for the 13 OECD countries that are analysed: At one extreme, only about one in three French and Portuguese workers are re‑employed within one year of displacement, whereas nearly 90% of displaced workers are re‑employed within a year in Finland and Sweden. Countries with low first year re‑employment rates make up some of the ground in the second year following displacement, but international differences still remain large as regards the probability of returning to work.15 Re‑employment rates for the peak crisis years (2009‑10) and the early post‑crisis recovery are somewhat lower than the pre‑crisis rates, while international differences in the speed of re‑employment remain quite stable across the three periods considered.16 OECD (2013[2]) shows that re‑employment rates in almost all countries are below‑average for women, workers with less than a secondary education and, especially, for older workers (aged 55‑64 years). The time spent out of work after displacement reflects a combination of unemployment (i.e. active but unsuccessful job search) and labour force withdrawal and is one of the main sources of lost earnings due to displacement, particularly in the first year or two following a dismissal.

Figure 4.4 shows regression‑based estimates of mean earnings impacts of displacement, which are inferred from differences between the evolution of the earnings for workers who are displaced in a given year and a control group of workers who were not displaced in that year.17 In all of the countries analysed, earnings fall during the year of displacement (DY), sometimes quite sharply (e.g. by nearly 50% in Portugal).18 The earnings gap created by displacement then declines very substantially during the next three to four years, albeit without having fully disappeared by the end of the observation window. These earnings impacts represent the combined effect of the time required to find a new job, when earnings are zero, and any change in earnings or job stability between the former and new jobs. A number of national studies of the earnings impact of displacement have shown that both the time out of work and lower earnings on the post‑displacement job (due to both lower wages and lower hours) make important contributions to overall earnings losses and that these losses can be large and persistent for some displaced workers, even if they are modest on average after the first year or two.19 The negative consequences of displacement are not limited to reduced earnings. For example, a number of studies have also documented declines in the health status of workers who are displaced, including an increased incidence of depression, hospitalisation for alcohol‑related conditions and higher mortality, as surveyed by Bassanini and Caroli (2015[9]).

The mean impact of displacement on earnings, which is presented in Figure 4.4, hides the very large variation in the costs borne by different workers. Figure 4.5 illustrates the heterogeneity of displacement costs. Panel A shows that the economic cost of being displaced tends to increase quite sharply with age, although the age gradient is not equally steep in all countries. OECD (2013[4]) provides additional examples of differences in the average earnings impact of displacement that are associated with observable characteristics, showing that earnings losses generally decline with the level of education, whereas the size and sign of gender differences in earnings losses vary from country to country (e.g. the earnings losses following displacement are higher for women than for men in the United States, but the opposite is true in Finland).

Figure 4.3. Displaced workers find new jobs much more rapidly in some OECD countries than in others

Panel B of Figure 4.5 zeros in on the impact of displacement on wages once workers have become re‑employed. Based on data that were collected for four of the countries where policies to assist displaced workers were reviewed, it can be seen that wages on the new job range from being considerably above those on the old job to much lower. While the share of (re‑employed) displaced workers experiencing wage increases of 10% or more ranges from 11% in Japan (where the economy was quite stagnant) to 41% in Australia (where economic growth was strong), the shares experiencing substantial wage reductions are more uniform: around 30% experienced a wage cut of at least 10% and around 10% a wage cut of 30% or more.

Figure 4.4. The depth and persistence of the reduction in earnings following displacement varies considerably across OECD countries

The estimates presented above reflect relatively short‑run wage effects for workers who found a new job within a year of being displaced. However, a number of national studies have looked at wage effects after more time has elapsed and a higher share of displaced workers have found new jobs, generalising the finding that large wage losses are experienced by a sizeable minority of displaced workers, while also showing that these losses are quite persistent. A number of these studies have also shown a concentration of large and persistent earnings losses among older, long‑tenure workers, particularly when they are also blue‑collar workers lacking a tertiary education. The pre‑displacement wages of many workers with this profile tend to reflect, in considerable part, returns to specific human capital that are often lost when these workers are displaced.20 The polarisation of the labour market in recent years has probably worsened re‑employment prospects for experienced workers losing medium‑skill production jobs, since relatively few new job openings match well with even the more portable skills possessed by this group (OECD, 2017[10]).

Policy implications

These empirical findings help to define the challenge facing labour market programmes intended to reduce the adverse consequences of job displacement for workers. In particular, they confirm that:

Figure 4.5. The impact of job displacement on earnings is highly variable

Displacement is an important source of unemployment, earnings insecurity and other types of hardship for workers. Cost‑effective measures to reduce the adjustment costs borne by displaced workers could thus serve important policy goals.
The large cross‑country differences in the speed of re‑employment following displacement and the size of the earnings losses once re‑employed suggest that well designed labour market policies and institutions may be able to significantly lower the costs of displacement without undercutting labour market dynamism.
Providing effective adjustment assistance to displaced workers is complicated by the fact that this is a very heterogeneous group in terms of their personal characteristics and employment histories. Most importantly, the nature and size of the displacement‑related costs that they bear range from large and persistent for the hardest‑hit individuals to virtually non‑existent for workers who move quickly to a new job that is as good as or even better than the lost job. This heterogeneity suggests that how much and which types of assistance displaced workers require is likely to be highly variable.

4.1.2. Overview of measures for reducing the costs borne by displaced workers

Table 4.1 provides a taxonomy of the wide range of public and private measures that potentially can reduce the adjustment costs resulting from job displacement. In particular, the table differentiates between direct and indirect measures (the table columns), and general and targeted measures (the table rows). This taxonomy is intended to illustrate the need to coordinate the labour market measures, which are analysed in detail in this chapter, with other policies that affect the incidence and consequences of displacement. It also highlights several important policy design issues and provides a reminder that many of the policy levers that can be used to lower displacement costs have potentially high efficiency costs and should be used with care or avoided altogether.

The first column of Table 4.1 provides illustrative examples of the types of direct measures to assist displaced workers that were the focus of the nine Back to Work country reviews and which are analysed in the rest of this chapter. Direct measures include the core active and passive labour market programmes, such as job‑search assistance, training and unemployment benefits. Much of the public re‑employment and income support received by displaced workers is provided by general measures in the sense that these programmes do not specifically target displaced workers and also serve many other jobseekers. For example, unemployment benefits and job‑search assistance are typically available to most or all unemployed persons (and, sometimes, to some employed persons). Targeted measures that are specifically designed to assist displaced workers also play a role in all nine of the countries that have been reviewed, although their importance and design show considerable variation from country to country. While targeted measures have the potential to be tailored to the particular difficulties encountered by displaced workers, they can also create wasteful duplication of programmes and inequities in the access to adjustment assistance across jobseekers who face similar difficulties.

The last two items in the first column of Table 4.1 illustrate the important role that private actors, particularly employers, can play in limiting the adjustment costs borne by displaced workers. For example, employers providing workers advance notification of layoffs improve the chances of these workers to make a smooth transition to another job, by allowing them to get an early start at searching (or retraining) for a new job. Another example of employers contributing to a successful adjustment is when they offer outplacement services to workers they displace, possibly in collaboration with trade unions. The nine country case studies that this chapter draws upon provide a strong confirmation of the importance of constructive employer and union engagement in assisting displaced workers, especially when these private initiatives are effectively coordinated with public assistance for this group.21 However, they also highlight how difficult it can be for governments to foster effective employer engagement when it is not spontaneously offered. Difficult co‑ordination issues can also arise between public and private measures to assist displaced workers. As will be discussed in Sections 4.3 and 4.4, one important co‑ordination issue is how, if at all, the receipt of severance payments should affect the access of displaced workers to public unemployment benefits and active labour market programmes (ALMPs).

Types of measures	Direct	Indirect
General	Unemployment insurance (UI) and other income‑replacement benefits available to all unemployed workers under common rules.	Macroeconomic and structural policies conducive to strong growth and high employment.
	Active labour market programmes (ALMPs) available to all unemployed workers under common rules.	Framework conditions for efficient reallocation of labour in response to structural change (e.g. adjustment‑friendly EPL and housing policies conducive to geographic mobility).
	Public insurance schemes against unpaid compensation when employers declare bankruptcy without making provision to fully compensate workers.	Education and training policies that anticipate and meet emerging skill demands.
Targeted	Special adjustment assistance or income-replacement benefits available to all displaced workers or to sub-groups of displaced workers (e.g. job losers in specific sectors or workers who lost their job due to a particular natural disaster).	Industry redevelopment or rationalisation programmes.
	EPL rules regulating economic layoffs, such as requirements for advance notification and severance payments, or rules about which workers are selected to be dismissed during a partial layoff.	Local economic development policies (e.g. geographically targeted tax or hiring subsidies, or public‑private partnerships to develop new sources of comparative advantage).
	Private outplacement services that employers and/or trade unions offer to certain displaced workers.	Trade policy measures to restrict imports such as tariffs and industry‑specific trade safeguards or anti‑dumping measures under WTO rules.

Note: EPL: Employment protection legislation. WTO: World Trade Organization. Several of the policy options included in the table are not recommended by the OECD because they are likely to do more harm than good (e.g. overly strict EPL and trade protectionism). They are included because governments have sometimes made use of these measures to protect workers at risk of displacement.

Many public policies have important, albeit indirect, impacts on how well displaced workers fare. The second column of Table 4.1 provides a few examples of the large number of indirect measures that potentially could reduce the adjustment costs borne by displaced workers. These include: i) macroeconomic and structural policies that are conducive to high employment and labour mobility; ii) educational and vocational training policies that improve the overall employability of mid‑career workers, including their capacity to retrain should they experience displacement; iii) local economic development policies that stimulate job creation in areas affected by mass layoffs and iv) housing policies that are conducive to geographic mobility. While this chapter does not analyse these framework conditions in any detail, it should be borne in mind that the direct measures to assist displaced workers, which are analysed here, are only one component of the broader policy strategy that is required to manage economic restructuring and its potentially disruptive impact on workers.22 Consideration of the types of indirect measures that are sometimes advocated to reduce the costs borne by displaced workers also highlights the risk that policy responses that undermine economic dynamism, such as trade protectionism, may be adopted if economically efficient measures to assist displaced workers are not put in place.23

Rigorous evaluation evidence on what works for displaced workers is surprisingly sparse

Much is now known about the effectiveness of different types of active labour market programmes (ALMPs), thanks to the growing number of rigorous evaluation studies that have been conducted, particularly in Western Europe and North America (Card, Kluve and Weber, 2015[11]; Kluve, 2010[12]). One common finding from these studies is that the effectiveness of any given type of re‑employment support is greater for some groups of jobseekers than for others, underlining the importance of tailoring the services offered to each jobseeker to their particular needs. Unfortunately, relatively few evaluation studies have examined the effectiveness of ALMPs specifically for displaced workers, leaving considerable uncertainty about how informative existing evaluation results are for assessing what works for the group that is the focus of this chapter, namely, formerly stable workers who experience an economic dismissal.24 There is also the added complication that displaced workers are a very heterogeneous group. For example, the types of measures that are cost‑effective for a younger displaced worker with good opportunities for labour market mobility are very likely to differ from the types of assistance that would be most cost‑effective for an older long‑tenure displaced worker whose largely manual skills do not match up well with the current structure of labour demand.

There are a modest number evaluation studies that have estimated programme impacts for displaced workers and which provide useful information about what works for this group. However, they are as yet too few to judge whether their findings generalise beyond the specific programmes and countries that have been studied. Notable examples include:

The largest number of evaluation studies singling out displaced workers have been conducted in the United States, where the Federal government has targeted funding to programmes that serve this group: both the “dislocated workerˮ track of the main ALMP funding stream provided by the US Department of Labor and the Trade Adjustment Assistance (TAA) programme for trade‑displaced workers. Overall, the evaluation results for measures to assist displaced workers have been very poor and cannot be said to provide consistent evidence that any of the main types of interventions studied have been cost effective.25 However, there are several reasons to think that these disappointing results may provide an excessively pessimistic picture of the potential effectiveness of ALMPs for displaced workers. First, many of the workers receiving job search, retraining and other assistance in these evaluation studies had already been unemployed for an extended period of time before receiving the assistance that was evaluated, whereas re‑employment assistance is likely to be more effective if initiated early in the unemployment spell or even during the notice period. A second reason is that the fragmented nature of the US system of labour market programmes and vocational training means that many members in the control group of “unservedˮ displaced workers actually received a similar service from another source, causing the estimated impact to understate the difference between receiving the service being evaluated (e.g. retraining) and not receiving it from any source. It is also possible that the relatively limited re‑employment impacts of the active measures evaluated reflects, in part, how these services generally were not integrated into a broader activation strategy.26
OECD (2015[13]) highlights key results from several recent evaluations of various types of ALMP measures that were offered to displaced workers in Canada. Quite consistent evidence is found that job‑search assistance, targeted wage subsidies and training all increased post‑displacement participation, employment and earnings, with the benefits from job search assistance and targeted wage subsidies being larger than those from training. A particularly encouraging result is that older displaced workers benefited even more from most of these measures than their younger counterparts. Another important finding is that post‑programme gains in both employment and earnings were larger when the services were offered more quickly following displacement.
Several evaluation studies in Europe also suggest that ALMPs can reduce the costs borne by displaced workers. In France, displaced workers signing a contract of professional security (contrat de sécurisation professionnelle, frequently abbreviated as CSP) are able to access more personalised and intensive re‑employment and retraining assistance from the public employment service than is generally available to unemployment insurance benefit recipients. Evaluations have confirmed that displaced workers who sign a CSP receive more intensive assistance than other jobseekers and that this group finds a new job more quickly – and has a higher probability of finding a relatively stable job – than other jobseekers with a similar profile who receive the regular PES services, DARES (2017[14]). A random assignment evaluation of the rapid provision of intensive counselling and job‑search assistance for older jobseekers in Switzerland showed that these measures significantly increase re‑employment for participants between the ages of 45 and 54 years, but had no effect for participants age 55 or older (Arni, 2012[15]). Intensive job‑search counselling appeared to promote re‑employment, in part, by convincing displaced workers to adopt more effective job‑search strategies, including more realistic wage targets for the new job.

Two additional limitations of the ALMP evaluation literature for the purposes of this chapter, are especially notable. First, few or no studies could be located that evaluate the effectiveness of several types of re‑employment assistance that the Back to Work country reviews identified as appearing to be particularly effective. In particular, rigorous evaluations of early intervention measures, which are the subject of Section 4.2, appear to be almost completely lacking. Indeed, it is difficult to construct a control group or another type of counterfactual for workers who benefit from these measures, since they typically provide re‑employment services and counselling to all of the workers affected by a mass layoff, beginning during the notice period.27 A second limitation is that there appears to be no evidence about the effectiveness, in terms of speeding up the re‑employment of displaced workers, of the behavioural requirements that many OECD countries include among the eligibility rules for unemployment benefits.28 It appears plausible that the subset of displaced workers with the best re‑employment opportunities might be particularly likely to hasten their job search in order to avoid participating in activities that they would be unlikely to value very highly or finding themselves subject to benefit sanctions. However, there does not appear to be any evidence about whether this group really is particularly responsive to the so‑called “threat effect” (Filges and Hansen, 2017[16]).

While it is to be hoped that additional rigorous evaluation evidence on the effectiveness of different types of labour market measures for displaced workers will be forthcoming, the more descriptive and qualitative evidence collected in the nine Back to Work country reviews already provides much useful guidance concerning potentially effective policy actions to lower displacement costs. The general lessons that emerge are summarised in Sections 4.2 to 4.4. It should be borne in mind, however, that many of the policy approaches that are identified as promising have not been rigorously evaluated.

4.2. Prevention and early intervention measures

One important difference between displaced workers and most other recipients of public employment services and unemployment benefits is that there is greater scope for proactive measures in the case of displaced workers. This is particularly true when employers provide workers and public labour market authorities with a significant amount of advance notice of layoffs, opening up the possibility of taking actions to save jobs that are still economically viable or to help workers to begin the adjustment process before they become unemployed. The columns of Table 4.2 identify the main types of proactive public policies and private initiatives that potentially can lower the costs borne by displaced workers, while the rows indicate whether the potential benefits take the form of avoiding socially‑inefficient layoffs or of speeding job search and re‑employment for workers whose jobs are no longer economically viable and thus should not be saved.

	Types of measures
Policy goals	Employment protection rules applying to permanent workers and mass layoffs	Experience rated financing of UI benefits	Short‑time work schemes	Private outplacement services (prior to layoff)	PES early interventions services (especially rapid response services for mass layoffs)
Prevention: Preserving viable jobs	X	X	X
Early intervention: Minimising post‑displacement adjustment costs by getting an early start on finding a new job	X			X	X

Note: X denotes a major policy goal of the indicated policy measure. PES: Public employment service. UI: Unemployment insurance. Employment protection legislation (EPL) is intended to promote both prevention and early intervention. The EPL components that are most relevant for prevention include those that effectively tax employers for displacing workers and certain procedural requirements (e.g. to consult with workers, unions or public authorities about alternatives to layoffs). EPL requirements for employers to provide advanced notification of layoffs and outplacement assistance to the affected workers are particularly relevant for expanding early intervention measures. While EPL can potentially promote prevention and early intervention, it can also create high efficiency costs.

The main policy measures intended to preserve jobs that pass a social benefit‑cost comparison are those that either: i) effectively tax layoffs (e.g. EPL rules that impose costs on employers who dismiss permanent workers and the experience rating of employers’ contributions to the UI benefit system); or ii) subsidise firms to preserve jobs, which are viable in the long run, during a short period when the employer has less or no need for those workers (e.g. short‑time work schemes). As regards early intervention measures that assist displaced workers to get an early start at finding a new job, the types of assistance that are offered are broadly similar to the ALMPs that public employment offices provide to unemployed jobseekers. However, there is some customisation of the content and organisation of the re‑employment and retraining services that are delivered to workers who are still employed but on notice that they soon will be dismissed. Another difference is that employer‑provided assistance plays a larger role at this early stage of the post‑displacement adjustment process, albeit more so in some countries than in others.

The rest of this section discusses job preservation and early intervention measures in more detail, focussing on the practical issues that arise in making effective use of these tools. Good practice examples are provided in both domains, along with several pitfalls to avoid. In particular, the important limits to the use of both preventive and early intervention measures is emphasised: Overuse of prevention measures impedes efficiency‑enhancing labour reallocation while trying to provide too much re‑employment assistance prior to displacement can disrupt production, potentially discouraging employer collaboration in early intervention measures, or wasting resources by providing costly re‑employment services to workers who are able to find a suitable new job on their own (or largely on their own).

4.2.1. Preventing inefficient layoffs without hampering the creative destruction process

Standard economic theory suggests that displacement rates would be likely to exceed the social optimum in the absence of policies that cause employers to take account of the social externalities associated with layoffs, such as the need to finance the unemployment benefits that will be paid to workers who are displaced (Blanchard and Tirole, 2008[17]). Employment protection legislation (EPL) is the policy instrument that is most commonly used to limit overuse of economic dismissals. Mandatory severance payments and certain procedural requirements (e.g. an obligation to provide re‑employment assistance to workers who are dismissed or to reinstate workers who successfully challenge their layoff in a labour court) are particularly likely to increase employer‑borne costs associated with job displacement and thus to discourage layoffs when the associated economic gains for the firm would be quite small in the absence of EPL. However, experience has shown that EPL needs to be used cautiously because it has often hindered efficiency‑enhancing labour mobility (OECD, 2013[2]). A particular risk is that EPL worsens the re‑employment prospects of displaced workers and voluntary job changers, because it causes employers to be more cautious about hiring. A light touch is thus necessary with employment protection measures, placing the emphasis on provisions, such as mandatory advance notification, that facilitate prompt access to re‑employment assistance for displaced workers (see the discussion of early interventions measures below), rather than measures that only make it cumbersome or expensive for employers to reduce staffing levels through dismissals.

Unemployment insurance (UI) schemes can be structured so as to discourage excessive layoffs, either through the experience rating of employers’ UI contributions or through a short‑time work (STW) scheme. Economists have often advocated experience rating as providing a more efficient instrument for forcing employers to internalise the social costs of job displacement than EPL, because it operates as a straightforward tax on layoffs that is tied to an important component of the social costs associated with displacement – see e.g. Albrecht and Vroman (1999[18]); Cahuc and Malherbet (2004[19]). Experience rating has not been widely adopted in practice, but may have considerable potential to discourage excessive layoffs without impeding desirable labour mobility, either on its own or in combination with a STW scheme (Cahuc and Nevoux, 2017[20]).29 While the evaluation evidence for experience rating is largely limited to the United States and provides mixed results, recent experience suggests that well‑designed STW schemes are able to preserve significant numbers of viable jobs during recessions without creating large efficiency costs. This potential is illustrated by the effectiveness of the Kurzarbeit programme in Germany during the 2008‑09 global crisis (Boeri and Bruecker, 2011[21]). Among the nine countries participating in the Back to Work reviews, the experience of Japan following the global financial and economic crisis provides a good example of how STW can help to preserve jobs during a deep but relatively short recession. In both Japan and Germany, enrolment in the STW scheme expanded rapidly in 2009, when business conditions worsened (and many employers may have faced financial constraints making it difficult to preserve viable jobs), but quickly returned to very low levels as the economy began to recover. By contrast, enrolment in the Finnish STW programme has remained quite high in recent years, even during economic expansions, probably because employers do not bear any of the cost of providing STW benefits.30 This appears to be a deliberate policy choice in Finland, where STW serves, in part, as an implicit subsidy for sectors where employment is highly seasonal. However, there is a risk that STW may be overused by Finnish employers, with the result that it ends up subsidising jobs that are no longer viable and which end, in any case, once the subsidy payments expire.

In sum, there is some potential scope for prevention measures to reduce displacement costs. However, it is limited and governments need to guard against the danger that excessive recourse to prevention measures creates high efficiency costs by impeding the reallocation of workers toward more productive employments. As regards reducing the costs borne by displaced workers, the main policy focus needs to be on measures to improve the re‑employment outcomes of workers whose jobs are no longer economically viable, while also compensating them for some part of the earnings losses that cannot be avoided. Since most of the emphasis should be placed on the promotion of successful job mobility after displacement, it is important to ensure that re‑employment measures are as effective as possible, including by initiating them as early as possible.

4.2.2. Early intervention measures

Early‑career job transitions are usually a source of career advancement (see Chapter 6). By contrast, mid‑career job mobility can be a difficult and time‑consuming process, especially when the job transition is involuntary and it affects long‑tenure workers who have not searched for a job on the external labour market in many years and have a strong emotional attachment to the type of work they are familiar with. Getting an early start on making this transition can be advantageous for a number of reasons. Even if the total time to become re‑employed remains unchanged, beginning during the notice period reduces the amount of time spent out of work and, hence, the earnings losses associated with displacement.31 However, it may also be possible to speed up the adjustment process and achieve better outcomes by starting the process before workers become unemployed. For example, prospective employers may tend to view job applicants who are still employed more favourably than those who are unemployed and it is well known that the longer a worker is unemployed the more their labour market prospects tend to deteriorate.

Other advantages to early interventions can be cited, particularly in the case of mass layoffs. For example, group counselling and job‑search orientation activities can be more easily organised during the notice period, particularly if the employer allows these services to be delivered at the work site.32 Group activities have two advantages in this context. First, they make it easier to meet the sudden increase in the local demand for re‑employment services that can easily overwhelm the service capacity of the local PES, especially when all services are delivered on an individual basis. Group activities can also be more effective in some cases. Since many of the workers affected by a mass layoff are confronting the same issues, group activities can be useful psychologically, for example in helping to overcome the reluctance of many experienced workers in a declining sector or occupation to consider possible career shifts, while also offering good opportunities to bring workers, who are about to be displaced, together with potential employers (“job fairs”).

Apart from the greater emphasis on group activities, the substantive content of the early intervention measures that are provided by the PES appear to be quite similar to the re‑employment services that are offered to newly registered unemployment benefit recipients. There is typically a focus on orientation activities, such as informing workers of the types of assistance that are available to them and how to access them. Another common focus is to assist workers to develop realistic strategies to find a new job, taking into account their skills and interests, labour market conditions, and the fact that many of them lack recent experience in searching for a job. While intensive measures, such as training, typically do not start during the notice period, considerable attention is often devoted to documenting workers’ skills and assessing how they align with employment opportunities in the local and national economy, including whether they should consider retraining.33

While there is little hard evidence quantifying the advantages of early intervention measures, the country visits undertaken in connection with the nine Back to Work reviews made it clear that practitioners believe that the benefits are considerable and they invest a lot of energy and resources in providing what are sometimes referred to as “rapid response servicesˮ.34 However, there is also a lot of variation across OECD countries in the way that early interventions are organised and the barriers they encounter in providing timely services to all of the displaced workers who would benefit from them. While the approach adopted needs to vary depending on national circumstances, such as how actively social partners participate in assisting displaced workers, it is nonetheless instructive to consider the main organisational issues that arise and the strengths and weaknesses of the different approaches that have been used to address them.

How large a role for employers and unions?

The extent to which employers and unions are actively involved in planning and providing re‑employment services to displaced workers is highly variable (see also Chapter 3) and can have an important effect on the feasibility and effectiveness of early interventions, as well as on the most effective way to organise public early intervention measures. This diversity is best illustrated by considering a few examples:

The Rapid Re‑employment and Training Service (RRTS) in Ontario provides re‑employment services for workers affected by larger scale layoffs, beginning during the notice period (see Box 4.1 for a more detailed description). It provides a good example of the public provision of early intervention measures where Employment Ontario (the provincial PES) plays a leading role, but other government agencies are also mobilised, notably Ontario’s Ministry of Training, Colleges and Universities. The RRTS illustrates the importance of adapting the level and type of service offered to the severity of the layoff and the capacity of the local employment service providers.35 For example, a temporary office (an Action Centre) is set up at or near the work site, when a large layoff threatens to overwhelm local re‑employment services.

Box 4.1. The Rapid Re employment and Training Service in Ontario, Canada

The Rapid Re‑employment and Training Service (RRTS) in Ontario provides an immediate response to large‑scale layoffs with the objective of connecting individuals with Employment Ontario (PES) services to help them regain employment. The level and type of support offered is tailored to the severity of the layoff and the capacity of the local employment service providers, which are typically third‑party providers, such as non‑profit firms, that the PES has engaged to provide employment services in a particular locality. The type of RRTS services offered may vary from:

Tier 1: If local re‑employment services have sufficient capacity to assist the affected workers, then the RRTS is limited to delivering information sessions and raising awareness of the employment services available to workers who will lose their jobs. These sessions may take place at the local Service Canada office (the agency where workers access Employment Insurance benefits) or by arranging for Employment Ontario service providers to go on site or extend their hours of operation so that affected workers can access their services before layoffs occur. Anecdotal evidence suggests that around 90% of the layoffs are dealt with using Tier 1 service, but these tend to be the layoffs affecting relatively few workers.

Tier 2: In a situation where nearby Employment Ontario service providers do not have sufficient capacity to meet demand. Supplementary funding is provided for outreach to affected workers, sometimes including the establishment of an Action Centre to deal with large scale closures. Through these Action Centres, displaced workers can access: i) job‑search assistance; ii) financial counselling and personal counselling to deal with the stress of job loss; iii) individual and group needs assessment; iv) vocational and educational counselling; and v) referral to programmes of Employment Ontario including the Second Career programme for older displaced workers. Every laid‑off worker develops an action plan within 15 days of his or her initial assessment and can access customised training, skills upgrading, job placement and relocation services. In general, Action Centres should not operate for more than a year.

Tier 3: When displacement occurs on a sufficient scale to have an adverse impact on the local economy or the community, a larger and a broader inter‑ministerial approach is taken. A local adjustment committee is established, which is led by an independent chair and is responsible for co‑ordinating the implementation of the Service Action Plan. This plan is put in place within 30 days of the initial response and outlines the roles of each of the service providers in the community who will be delivering services to displaced workers. A key component of the process is the development of a multi‑disciplinary Rapid Response Team, which is formed at the local or regional level to provide timely, focused and integrated training and employment solutions to affected workers and communities.

Source: OECD (2015[13]), Back to Work: Canada – Improving the Re‑employment Prospects of Displaced Workers, http://dx.doi.org/10.1787/9789264233454-en.

Job Security Councils (JSCs) in Sweden provide an example of re‑employment services for displaced workers, including early interventions, which are provided by the social partners (i.e. employer federations in close collaboration with union federations), rather than by the PES or other public actors (see Box 4.2 for a more detailed description). The JSCs make productive use of the comparatively long notice period in Sweden to speed the return to work following displacement and are probably a key explanation why nearly 90% of workers are re‑employed within a year of being laid‑off. This model is particularly successful at extending early intervention measures to individual and small‑scale layoffs. One weakness of the JSC system is that there is considerable variation in the level of support the councils provide to different displaced workers.
Hybrid models that feature a more even mix of public and private actions to promote successful adjustment can also be found in OECD countries. For example, the Change Security programme in Finland represents a close partnership between the PES, employers and unions in managing displacements (OECD, 2016[22]). Even in countries where the involvement of employers in the provision of re‑employment assistance to displaced workers is not strong in general, some employers voluntarily provide outplacement services to workers that they are displacing and these services can represent an important supplement to the services provided by the PES. For example, (OECD, 2015[23]) discusses how large Japanese employers who displace regular workers typically engage private placement firms to assist the affected workers in their search for a new job. While there is no legal obligation for employers to provide outplacement services in Japan, doing so appears to be closely related to the human resources strategies associated with the “lifetime employmentˮ system, which is characterised by a strong employer commitment to providing a high level of employment security to regular workers.36

Can governments foster greater employer engagement?

Since constructive employer engagement is of great value in providing prompt and effective re‑employment assistance to displaced workers, it is important to assess what governments can do to promote such engagement where it is not already well rooted in national industrial relations institutions and practices.37 OECD governments make use of two main policy strategies to foster greater employer engagement than is spontaneously offered: i) legal compulsion via EPL; and ii) a softer touch approach relying on incentives and moral suasion. These two approaches can also be used in combination. For example, a requirement in Quebec province in Canada that employers organise outplacement services for displaced workers in the form of an outplacement assistance committee (Comité d’aide au reclassement, or CAR) is combined with a 50 % subsidy of the cost of those services (OECD, 2015[13]).

There is a role for EPL rules to require some minimal level of constructive employer engagement in improving the re‑employment prospects of workers they displace. In particular, there appears to be a clear case to require at least a minimum level of advance notification to workers, unions and public labour market authorities, since it is a sine qua non for early intervention. Figure 4.6 shows that advance notice requirements differ strikingly across OECD countries, ranging from no general requirement in the United States38 to quite long notice periods in some European countries. Consistent with the findings that displacement costs rise strongly with job tenure, many countries require longer notice periods for workers with greater tenure. In some countries, notice periods also vary between occupational groups, but it is questionable whether there is a good economic rationale for those differences.39

Box 4.2. Job Security Councils in Sweden

Job Security Councils (JSCs) were first developed in the 1970s against the backdrop of the deteriorating economic conditions in Sweden in the late 1960s and the massive job losses among white‑collar workers in the wake of the oil crisis in 1973. In this context, the public employment service was not regarded by employers as providing sufficient support for white‑collar workers to find new jobs (Diedrich and Bergström, 2006[24]).

The councils are based on collective agreements between the social partners in a sector or occupational field, such as white‑collar workers in the private sector. JSCs are actively involved in all stages of the process of restructuring, including by providing advice to employers and trade unions at an early stage in the process. They also provide transition services and guidance to workers who are made redundant, through individual counselling, career planning, job‑search assistance, outplacement services and retraining. The councils’ activities are financed by employer contributions which are fixed as a percentage of their total payroll. The contribution level is determined as part of the collective agreement (currently 0.3% of payroll). As such, the council operates as a form of insurance, spreading the costs of restructuring across all employers who are covered by the collective agreement. A particular strength of the JSCs is that prompt re‑employment support is offered to all displaced workers covered by the agreement, including workers in small and medium‑sized enterprises (European Commission, 2010[25]).

JSCs appear to make a significant contribution to the unusually rapid re‑employment rate of displaced workers in Sweden (cf. Figure 4.3). Indeed, the councils report that around 90% of their participants find a solution within nine months: 78% finding new employment, 8% starting a new business and 6% enrolling in longer duration education or training. This success reflects a strongly proactive orientation, a comparatively long notice period, and JSC staff’s intimate knowledge of the workers they serve and their occupational labour markets. However, despite the overall very positive experience with JSCs in Sweden, heavy reliance on the councils to provide re‑employment services also raises several concerns. One concern is that the private‑sector, white‑collar JSC offers considerably more intensive re‑employment and, especially, retraining services then the main blue‑collar JSC, probably because the former has considerably greater resources per displaced workers. These differences in the level of support offered are reflected in re‑employment outcomes: whereas 65% of the white collar workers serviced by their JSC find a new permanent job within six months, this is the case for only 38% of their blue‑collar counterparts. Another source of unequal access to re‑employment support is that approximately 20% of the workforce is not covered by a JSC. Finally, coordination between the JSCs and the PES is quite limited, making it difficult for the PES to complement the services offered by the JSC in a timely manner.

Source: OECD (2015[26]), Back to Work: Sweden: Improving the Re‑employment Prospects of Displaced Workers, http://dx.doi.org/10.1787/9789264246812-en; Diedrich, and Bergström (2006[24]), “The job security councils in Swedenˮ, http://imit.se/wp-content/uploads/2016/02/2007_145.pdf; and European Commission (2010[25]), “27 National seminars on anticipating and managing restructuringˮ, http://www.employment-studies.co.uk/resource/27-national-seminars-anticipating-and-managing-restructuring-arenas.

Figure 4.6. Legally mandated notice periods vary widely

Unfortunately, there is little evidence that can be drawn upon to identify the optimal level of notice. In particular, it is not clear when increases in the length of the notice period begin to translate into additional benefits for displaced workers, in terms of easier transitions to new jobs, that are too small to justify the additional costs that result for employers (e.g. in terms of disruptions to production due to low worker morale or increased wariness of credit markets, suppliers and customers to engage with a firm seen to be struggling). Nonetheless, countries with relatively low notice requirements should consider raising them closer to OECD averages, perhaps in combination with relaxing other EPL requirements, such as the level of mandatory severance payments. Another open question is whether advance notification requirements should be combined with an obligation for employers to offer workers a minimum number of days of paid job search leave during the notice period. For example, workers on notice of displacement are entitled to between 5 and 20 days of paid job search leave in Finland, depending on the length of their notice period (OECD, 2016[22]). However, little is known about how much these entitlements contribute to good re‑employment outcomes nor about the resulting cost for employers.

EPL requirements that employers provide re‑employment services to workers they displace may also be useful in some cases, although experience suggests caution because these types of obligations can impede efficiency‑enhancing labour mobility by imposing excessively high layoff costs on employers. The recent experience of France is informative in this respect (OECD, 2015[13]). Until recently, all medium and large employers were required to develop a job preservation plan (plan de sauvegarde de l’emploi, PSE or, more colloquially, “social planˮ) when displacing ten or more workers. These plans need to both specify the measures being taken to avoid as many layoffs as possible, as well as the severance payments and re‑employment services the employer will provide to workers whose jobs cannot be saved (often delivered via an ad‑hoc local team – cellule de reclassement). These plans are submitted to the work council for its review and can also be reviewed and possibly rejected by labour courts, if they are judged to be inadequate. This approach to managing layoffs is widely seen as expensive and overly complex for smaller employers, and it probably contributes to the heavy use of temporary employment contracts (i.e. as a means of avoiding these requirements when reducing employment levels). The process was also often conflictual, delaying the access of displaced workers to re‑employment services. Smaller and medium‑sized firms were also seen to lack the necessary expertise to organise effective re‑employment services. A series of reforms in recent years has shifted towards a new system for providing prompt and intensive re‑employment services to displaced workers, in which the PES plays the leading role in providing those services. Since 2011, workers in firms with fewer than 1 000 employees who are notified that they will be displaced can opt for a career path security contract (contrat de sécurisation professionnelle or CSP).40 The CSP entitles them to higher than usual unemployment benefits and rapid access to intensive re‑employment services from the PES, while releasing their employer from the obligation to set up a social plan. The cost of this programme is shared by the PES, the employer and the dismissed worker, providing incentives for all actors to facilitate better co‑operation and potentially improve the take‑up of the programme. Initial evaluations suggest that CSP signatories have greater re‑employment success than similar jobseekers who do not sign a CSP (DARES analyses, 2017[14]).

In countries where there are important gaps in constructive employer engagement, it is sometimes possible to encourage broader voluntary engagement through either subsidies or outreach policies. For example, the Labour Mobility Subsidy in Japan reimburses part of the costs incurred by employers who contract with a private placement firm to provide re‑employment support for workers they are displacing, provided those workers are placed into new jobs sufficiently rapidly (OECD, 2015[23]).41 The PES in the US states of Michigan and Pennsylvania provide examples of outreach policies. These states operate “early warningˮ systems to try to identify upcoming plant closings and other mass layoffs, for example by reading the business press and talking with various economic actors (OECD, 2016[27]). If they believe that a firm may be preparing a mass dismissal, they contact it to verify whether that is the case. If a pending layoff is confirmed, then the employer is encouraged to make use of the government’s ability to provide re‑employment services for its workers, for example by setting up a rapid response plan for them. While these efforts are worthwhile, their effectiveness is undercut when advance notice is often not provided or many employers prefer not to co‑operate with the PES. In the worst cases, “run‑away firmsˮ close or move out of the country without providing any notice or making any other provision for their workers.

Unequal access to early intervention measures

Another challenge affecting the provision of early intervention measures is that only a relatively small subset of displaced workers have access to these services in most countries: primarily workers who are laid‑off as a part of a mass layoff at a large firm, which triggers public rapid response services, or whose employer makes an effort to ensure that good re‑employment services are available. One advantage of the Job Security Councils in Sweden is their universal coverage of displaced workers in the sectors where they operate.42 Quebec province in Canada makes use of an interesting strategy to extend rapid response services to workers affected by small or individual layoffs. Displaced workers whose employer is laying off fewer than 50 workers and hence is not obligated to set‑up an outplacement programme of its own (i.e. a CAR), can enrol instead in a continuous enrolment outplacement plan run by the PES (comités d’aide au reclassement à entrées continues, CREC). Although the CREC have yet to be subjected to a rigorous evaluation of its effectiveness, it provides an interesting model for expanding access to early intervention services.

Displacement costs are probably above‑average for workers who are displaced by mass layoffs that result in chronic excess labour supply in the local labour market.43 That raises the possibility that the de facto concentration of early intervention measures on workers who lose their jobs during a mass layoff, as is the case in most of the countries reviewed, may tend to target additional re‑employment assistance to a subgroup of displaced workers who face particularly large barriers to successful job search and thus be a reasonably good way to direct limited budgetary resources. While there is probably some truth to this conjecture, the size of a layoff is likely to be an imprecise indicator of the adjustment difficulties faced by individual displaced workers.44 Even in instances when a mass layoff has a large ripple effect on the local economy that worsens re‑employment opportunities for displaced workers, many of the affected workers are likely to have lost their jobs as a result of small layoffs (e.g. at subcontractor firms or other local businesses), rather than in the initial mass layoff. This suggests that governments should attempt to extend access to early intervention measures to workers affected by small or even individual displacements, whenever cost‑effective ways can be found to do so. Resources permitting, it also seems best to provide early intervention services to all displaced workers, rather than only those facing above‑average re‑employment difficulties. Quite apart from how much re‑employment assistance a particular displaced worker requires, it usually will be an advantage to them to be able to access that assistance as soon as they are notified their job is ending, rather than needing to wait until they have become unemployed.

Co‑ordination challenges

Early intervention measures, particularly the establishment of rapid response services in anticipation of mass layoffs, raise several coordination challenges for the governments. One concern is illustrated by Australian employers who sometimes complained in the past about having been contacted by multiple government agencies in an uncoordinated manner when they were preparing a mass layoff, with the result that they answer the same questions multiple times and sometimes receive conflicting information about how they should coordinate their planning with the government and what services are available to the employer or the workers who will be displaced (OECD, 2016[28]).45 When employers (or the social partners) provide substantial re‑employment services to displaced workers, another coordination issue arises for the PES, namely, to make sure that public re‑employment services complement the private services, avoiding both wasteful duplication and the risk that certain workers fall between the cracks.46 As much as possible, government agencies should coordinate their interactions with employers and private providers of re‑employment services. One way to do this is to establish a formal coordination process, as was done for a recent mass layoff at the Sharp Corporation in Nara Prefecture in Japan (see Box 4.3).

4.3. Re‑employment assistance including retraining

After the notice period has ended and workers have been displaced, much of the policy focus should remain on active measures to promote rapid re‑employment. This section analyses how best that can be done within a broader national activation strategy (i.e. a co‑ordinated system of monitoring, sanctions and employment services that promotes transition to employment). The experience of many OECD countries confirms that enforcing the obligation of unemployment benefit recipients (and some recipients of other income replacement benefits) to search actively for a job or participate in activities that raise their employability, while also providing them with the re‑employment supports that they need can significantly speed the transition into suitable jobs – see OECD (2013[2]; 2015[29]).47 It stands to reason that displaced workers would be more likely to benefit from effective re‑employment services in countries that operate a strong overall activation strategy. However, minimising the costs that workers bear following displacement also requires the general principles of activation policy to be applied to displaced workers in a way that addresses their specific needs for re‑employment support. This section analyses how best that can be done, drawing upon recent policy experience in OECD countries, particularly the nine countries that participated in the Back to Work reviews.

4.3.1. Overall spending on active labour market programmes

While the primary focus of this section will be on ensuring that displaced workers receive the right types of re‑employment support at the right time, it is useful to begin with a short review of overall spending on active labour market programmes (ALMPs) since this is likely to have an important effect on the services that displaced workers can access. Indeed, a number of cross‑country regression studies have concluded that higher ALMP spending is associated with better aggregate labour market outcomes and Andrews and Saia (2017[8]) recently extended that line of research by showing that re‑employment of displaced workers tends to be more rapid in countries with relatively high aggregate spending on ALMPs.

ALMP spending per unemployed persons as a percentage share of per capita GDP varies tremendously across OECD countries, including the nine countries participating in the Back to Work reviews. Panel A of Figure 4.7 shows that, in Denmark, spending on active measures for each unemployed person was as high as 64% of per capita GDP in 2015, by far the highest value observed in the OECD area. By contrast, in the United States, spending per unemployed was just 4% of per capita GDP, one of the lowest spending levels within the OECD.48

There are also striking cross-country differences in the way total spending is divided across the different types of programmes (Panel B of Figure 4.7). This heterogeneity also concerns the nine review countries.

Box 4.3. HQ Sharp in the Nara Prefecture

After a rapid deterioration of business conditions, Sharp announced an early retirement plan on 20 November 2012 that was intended to enrol 2 000 workers, but actually attracted 2 960 enrolees aged 40 and above. A considerable number of those workers lived in Nara Prefecture. In response to this announcement, the Nara Labour Bureau (the PES) and the Nara Prefectural Government jointly organised the Support for Sharp Related Displaced Employees Headquarters (“HQ Sharp”) in November 2012.

While the headquarters model has been used for other mass displacements in Japan, the composition of participating organisations and their mode of operation vary from case to case. Indeed, HQ Sharp was one of the best instances of a prefectural labour bureau and a prefectural government jointly establishing and managing headquarters' downsizing. This organisation was selected because it reflected the already close working relationship between the Labor Bureau and labour market programmes run by the prefectural government, such as its Job iCenter. The additional partners in HQ Sharp included the Industrial Employment Stabilization Center (IESC) in Nara – a private agency that facilitates employee transfers between participating firms – and four municipal governments.

HQ Sharp was a co‑ordinating committee consisting of managers from the participating organisations. Its mission was to build an integrated support system to offer effective re‑employment and livelihood aid for displaced workers. It was also intended to provide support measures for related businesses (e.g. suppliers for Sharp). Much of the work of the headquarters consisted of an extensive consultations process that was used to achieve agreement on the strategy to adopt and eventually the setting up and implementation of actions plans. Another key focus was to set up a system for exchanging relevant information. Finally, a system to provide vocational counselling at the job centre was set up.

The ultimate aim of HQ Sharp was to facilitate smooth transitions into re‑employment and to support the living standards of workers who lost their job. As part of achieving this goal, informational meetings were organised for workers who were to be displaced by Sharp. At these meetings, information was provided about various services that were available to these workers, as well as instructions about how to access these services. At these informational meetings, workers were also provided with temporary registration cards that they could fill out, if they wished to register for outplacement assistance from the Nara IESC.

Source: Information presented to the OECD Secretariat when it visited the Nara Labour Bureau, the Nara Prefectural Government, and the Nara office of the IESC in October 2013.

The largest share of spending went to basic case management and job‑search assistance (“PES and administration”) in Australia, Canada, Japan and New Zealand, while Finland concentrated spending on training. Denmark and the United States allocated large shares of their ALMP spending to both training and sheltered and support employment, while Sweden spent the most on employment incentives and Korea on direct job creation. It is rather remarkable that labour market practitioners in these nine countries described their efforts to support displaced workers back into work in a similar manner, when resource levels and spending priorities for ALMPs appear to be so different. That they do, suggests that many of the issues involved in providing the right services to displaced workers remain relevant at very different levels of overall resources, even if a resource‑rich environment is undoubtedly an advantage.

Figure 4.7. The re‑employment assistance available to jobseekers is influenced by overall spending on active labour market programmes

4.3.2. Do displaced workers receive the right types of re‑employment services?

How do the needs of displaced workers differ from those of other PES clients?

The empirical findings in Section 4.1 suggest that the job‑search aspirations and prospects of displaced workers differ significantly from the jobseekers who have been the primary focus of activation policies in many countries, where emphasis has often been placed on moving benefit recipients who are quite distant from the labour market into usually low‑skill and low‑pay employment (Immervoll and Scarpetta, 2012[30]).49 That raises the possibility that it may be difficult for employment services that have a strong focus on supporting relatively disadvantaged persons (e.g. low‑skilled individuals with little or no history of stable and reasonably well‑paid employment, including early school leavers and persons with partial disabilities) also to serve effectively mainstream displaced workers who had experienced considerable employment security in medium or well‑paying jobs prior to being laid‑off and hope to find new jobs of a similar quality. If both groups are to be served well, the specific support measures offered to displaced workers will need to differ in important respects from those offered to many of the other jobseekers supported by the employment service.

Immervoll and Scarpetta (2012[30]) and OECD (2015[29]) argue that an effective activation strategy needs to address three basic types of barriers to successful job search by taking measures to: i) strengthen the client’s motivation to look for and make use of existing earnings possibilities (e.g. by reinforcing work incentives and enforcing job‑search requirements, with benefit sanctions and warnings); ii) address labour‑supply side barriers to employment (e.g. by increasing employability with training and rehabilitation); and iii) expand earnings opportunities by connecting clients with suitable job openings or using demand‑side instruments, such as wage subsidies, to create employment opportunities. This taxonomy provides a useful framework for delineating the specific re‑employment barriers (and advantages) that are characteristic of displaced workers:

Motivation. Since displaced workers have a stable work history, they are usually characterised by strong labour force attachment and are highly motivated to return to work. An off‑setting factor is that many displaced workers qualify for relatively generous unemployment benefits and/or large severance payments and may thus be tempted to delay intense job search for a considerable period of time, if their search effort and work availability are not monitored effectively and they are not counselled about the risk that an extended period out of work is likely to be viewed as a bad signal by prospective employers. A history of stable and well‑paid employment can also generate overconfidence about how easily a new job can be found or unrealistic aspirations concerning the pay level or the possibility to remain in the same industry and occupation. However, excessive pessimism can also be a problem, particularly for older displaced workers or workers displaced from declining occupations who doubt their capacity to make a career transition. Early engagement with the employment service, including counselling about the adjustment process and encouragement to develop a realistic re‑employment strategy as quickly as possible, is thus likely to be especially valuable for displaced workers.
Employability. Familiarity with the world of work and a proven ability to perform on the job are typically not a problem for displaced workers. However, matching their skills with available jobs can be a challenge. This is particularly the case for older blue‑collar workers displaced from declining sectors and occupations. This group has typically acquired a lot of skills on the job that are not well documented and which may not match well with the more cognitive and social skills required in growing sectors and occupations. In such cases, skill audits that document the worker’s skills (sometimes referred to as recognition of prior learning or RPL) can be very useful, especially when combined with counselling that helps these workers to identify retraining strategies that can supplement their existing skills so as to qualify them for jobs currently in demand (“gap trainingˮ). Coaching in job‑search methods is also likely to be useful for long‑tenure workers who have not looked for a job in many years.
Opportunities. The public employment service may struggle to help place displaced workers who have lost good‑quality jobs, even if their occupational specialty is still in demand. Job placement is most effective when PES staff have a good knowledge of the relevant segment of the labour market and good contacts with employers. The widespread perception in the nine countries participating in the Back to Work reviews is that the PES is most effective at placing low‑skilled workers with relatively little or relatively unstable work histories into low‑paying jobs. This suggests that it may make sense for the PES to create a separate track of re‑employment services for more skilled displaced workers, possibly making use of private placement agencies that specialise in placing workers into higher paying jobs.50 When a large number of displacements in a region is associated with persistently depressed labour market conditions, the PES should also facilitate the geographic mobility of job losers who are receptive to the idea of moving to another region with a more buoyant labour market (e.g. through providing information about employment opportunities and subsidising moving costs). However, experience shows that many displaced workers (and their families) have strong ties to their community, implying that the focus often needs to be on promoting successful job placement where they live, possibly including measures such as hiring subsidies for local employers and broader efforts to diversify the local economy.51

Individual tailoring of re‑employment services for displaced workers

The empirical analysis in Section 4.1 showed that the labour market experience of displaced workers is incredibly varied: while a considerable number of displaced workers move quickly into new jobs that are about as good as or better than the lost job, another sizeable group incurs moderate income losses and a third group experiences a large and lasting decline in their earnings capacity, due to long‑term joblessness, large wage reductions on the post‑displacement job or a combination of the two. The heterogeneity in displacement costs complicates the provision of re‑employment services to displaced workers, since their individual needs for this type of support range from small or even non‑existent to large. Clearly, the goal should be to tailor the offered support to individual needs, so as to avoid both unnecessary spending on services for workers who can find a suitable new job with little or no assistance, on the one hand, and delaying access to intensive services to those who need them until they have been unemployed for a long time, on the other. However, this is more easily said than done.

Systematic early needs assessment for displaced workers represents the most straightforward approach to tailoring re‑employment services to match individual needs early in the unemployment spell. The focus would be to: i) better identify the jobseeker’s skills, relevant experience and opportunities in the labour market; ii) explore options for alternative career paths; iii) identify skills development needs and other barriers to re‑employment; and iv) refer to more intensive interventions (e.g. intensive individual counselling or training) only when specific barriers to re‑employment have been identified. This needs assessment would be the occasion for displaced workers to discuss their professional plans, retraining options and job‑search methods with a case worker. It could also be formalised in an individual action plan, at least in cases where significant barriers to quick re‑employment have been identified.

The Back to Work reviews show that current PES practice diverges sharply from such an approach. This could indicate that there is considerable scope for improvement. However, this divergence also suggests taking a cautious and incremental approach to introducing systematic early needs assessment for displaced workers, until it has been demonstrated that such an approach has been successful in practice. Among the considerations that arise, the following can be listed:

When a displaced worker registers at the PES, a profiling instrument and/or case worker judgement typically are used to make an initial assessment of individual needs and, thus, which re‑employment services the jobseeker can access early in their jobless spell. In making these determinations, it appears that little or no explicit attention is devoted to trying to differentiate between displaced workers who have good mobility prospects and those requiring more intensive assistance in any of the nine countries reviewed. Indeed, it is rare for the PES to classify its clients according to whether they were displaced or became jobless through another route.52 This probably reflects, at least in part, the finding in Section 4.1 that the labour market prospects of displaced workers are so varied. In effect, the statistical category that economic researchers have adopted for displaced workers, while useful for studying labour market turnover, is too broad to serve case workers as a useful proxy indicator of individual needs for re‑employment support.
It is possible that the case management practices that are used for all newly registered workers, whether or not they were displaced, implicitly capture the distinction between displaced workers who do and do not require intensive services, at least to some degree. While there is considerable variation in national practice, the logic for determining individual needs tends to be quite similar. Typically, information is collected on a number of factors thought to predict greater barriers to re‑employment (e.g. a long period out of work, poor skills or health problems). Whether or not this information is combined into an overall numerical score, it provides the basis for case workers judgements concerning the job seeker’s initial needs for re‑employment and retraining services.
Data on which re‑employment services displaced workers receive and when are generally lacking. Nonetheless, it seems likely that the needs assessment practices currently in use result in too few newly displaced workers being granted access to intensive re‑employment assistance early in the unemployment spell, although they eventually may be offered such services should they remain unemployed for a long enough period of time. This conjecture is based on the observation that newly displaced workers with a stable employment history do not fit the profiles that typically are used by PES offices to identify the persons at the highest risk of long‑term unemployment and benefit dependency.53
If displaced workers were assessed as a distinct group among new PES clients, would it be possible to better identify the individuals who would benefit from prompt access to intensive services? The spotty evidence available suggests that it is probably quite difficult to accurately assess the individual needs of newly displaced workers. For example, a random assignment evaluation of a pilot programme in Switzerland, which offers intensive re‑employment services to older displaced workers, concluded that the counsellors who worked closely with these jobseekers found it very difficult to anticipate which individuals were at greatest risk of long‑term unemployment (Arni, 2012[15]). This experience is consistent with the statistical evidence discussed in Section 4.1. While there are large average differences in displacement costs across groups defined by age, tenure and several other observable characteristics, much of the individual variation in costs remains even after controlling statistically for a considerable number of individual and job characteristics (cf. Figure 4.5).
While a general solution to individualising the re‑employment support that is offered to displaced workers is not yet available, the Back to Work reviews highlight how skills validation tied to training support is an area in which important progress has been achieved. The re‑employment prospects of displaced workers, particularly older workers whose vocational skills were largely learned on the job, can be greatly improved if their job skills are credibly assessed and documented in a way that makes it possible to ascertain how well they match up with skills credentials that are used in the external labour market. A number of OECD countries have developed recognition for prior learning (RPL) instruments that can be used for this purpose and the closing of a large Bridgestone tire factory in Adelaide in 2010 illustrates how effective RPL can be when the employer cooperates in documenting workers’ skills and re‑employment counsellors use the results of the RPL exercise to engage the worker in a discussion about whether retraining would be desirable and, if so, which type of training would most efficiently qualify that worker for a suitable new job (OECD, 2016[28]). The payoff to this approach is increased when vocational education and training providers are flexible about providing customized training courses that cover only the material that needs to be learned. Since employers who are recruiting new workers often place a lot of emphasis on work experience, as well as formal qualifications, the PES should also assess whether a temporary hiring subsidy should be used to make it easier for newly trained displaced workers to obtain some initial experience in their new occupation. Good practice examples of applying this general strategy were observed in many of the countries reviewed, but it appears that only a small share of displaced workers have access to these services.
In light of the difficulty of identifying which individual displaced workers would benefit most from quick access to intensive re‑employment services, it seems worthwhile to experiment with different approaches to identifying members of that group early in their unemployment spells. More targeting could also be done at the group level. In particular, greater access to intensive services could be offered to older long‑tenure displaced workers. As is illustrated by the Swiss pilot programme evaluated by Arni (2012[15]), the inability of case workers to forecast which older displaced workers in particular faced the greatest re‑employment barriers did not prevent the programme from speeding re‑employment and raising employment stability.54 Another group that could be offered greater access to intensive services at the beginning of their unemployment spell is displaced workers whose participation in early intervention measures, such as counselling and skills audits, reveals that they face important re‑employment barriers. There could also be some scope for self‑selection, such as limiting training access to displaced workers who have developed a credible proposal for retraining as part of a broader plan for career mobility.

4.3.3. Difficulties in providing displaced workers with prompt access to effective re‑employment services

Reaching displaced workers who do not immediately qualify for unemployment benefits

Access to even the most basic re‑employment services can be delayed when a considerable amount of time elapses between the time when a worker is displaced and the time when she becomes eligible to receive unemployment benefits. As is discussed in Section 4.4, eligibility rules for UI benefits in a number of OECD countries treat severance payments as compensation. For example, a severance award that is equivalent to six months of wages delays eligibility for unemployment benefits by six months in Australia, Canada, Finland and Sweden. This delay in receiving income benefits typically results in an equal delay in registration with the PES and thus of exposure to activation measures, such as job‑search requirements and counselling. In some cases, displaced workers are entitled to obtain basic job‑search assistance at employment offices even when they are not eligible for an income benefit, but these services tend to be quite limited and take‑up low. As regards ensuring displaced workers have prompt access to re‑employment services, the implication of these delays is that participation in re‑employment measures needs to be decoupled from the receipt of income support.55

Two strategies to decouple the initiation of re‑employment services from benefit eligibility are practiced by at least a few OECD countries:

Outreach. Some PES services are generally available to all workers, such as self‑service use of job search tools (e.g. an online database of job vacancies). Greater use of these resources could be encouraged among displaced workers who are not receiving an unemployment benefit by raising public awareness of the availability of these services and enhancing their value for users. While such measures would be potentially useful for labour market participants more broadly, it seems unlikely that such measures would be very effective in engaging many displaced workers who have received a large severance award. One way to more effectively reach displaced workers is illustrated by the Jobs and Training Compact that the Australian government introduced at the beginning of the global financial crisis and which temporarily allowed displaced workers to access an intermediate level of re‑employment support, rather than only basic services, even when they were not eligible to receive income benefits (OECD, 2016[28]).
Mandatory registration. In order to minimise unemployment duration and facilitate early contact with employment services, several OECD countries require workers to register with the PES as soon as they are notified that they will be dismissed, even though they are not yet eligible to receive unemployment benefits. For example, Switzerland requires displaced workers to give proof of job‑search activities between dismissal notification and the first interview at the PES to receive unemployment benefits (Duell et al., 2010[31]). A similar preventive approach was adopted in Germany as part of the Hartz reforms, with workers being obligated to register as jobseekers three months before their job ends or, for those with shorter notice, within three days of receiving notice of dismissal (Mosley, 2010[32]). This type of registration obligation allows the PES to make referrals of vacancies even before the first unemployment benefit payment. As much as possible, these sorts of requirements should be combined with the early initiation of at least basic re‑employment services, including during the notice period (cf. Section 4.2 discussion of early intervention measures).

Meeting sudden upsurges in the number of displaced workers

Another specificity of displaced workers is that the number of job losers requiring re‑employment services is quite variable and, in particular, is subject to sudden upward spikes that can overwhelm the capacity of ALMP providers to meet the increased demand. This is most frequent at the local level when one or several mass layoffs create an upsurge in the number of job losers requiring assistance, even as job‑search opportunities in the local labour market may worsen. Something similar occurs at the national level during a recession. Finally, natural disasters can cause widespread job displacement in the affected region which needs to be addressed, even as other urgent needs such as rescue, evacuation and rebuilding also require a vigorous response.

The nine Back to Work reviews, together with closely related OECD studies of the temporary expansions of ALMPs in response to the upsurge in displacements and unemployment that followed the global financial crisis (OECD, 2009[1]; 2010[33]; 2012[34]) and six recent natural disasters in OECD countries (Venn, 2012[35]), all suggest that labour market programmes have withstood these stress tests surprisingly well overall, albeit at the cost of increased spending and intense efforts by programme managers and the line staff to quickly put those extra resources to good use. Nonetheless, an improved capacity of labour market programmes to rapidly upscale re‑employment services for displaced workers remains a priority. One of the biggest challenges is to rapidly expand capacity without compromising quality. Another is to shift the mix of services that are delivered, so as to reflect changes in the composition of workers being served and labour market conditions. Several lessons can be drawn from recent experience:

There are important limits to how rapidly ALMPs can be up‑scaled at the national level, because spending levels typically rise much less than proportionally to the increase in unemployment during recessions and, even when increased funding is available, it is difficult to expand service supply quickly without diluting quality (e.g. it takes time to recruit and train case workers and other skilled staff). For example, ALMP expenditures per unemployed person fell quite sharply as unemployment surged following the global financial crisis, just as the share of displaced workers among the unemployed increased.56 However, the decline in ALMP spending per unemployed person was less sharp than would have been predicted based on spending patterns in earlier recessions, probably due to the increased priority governments have come to place on activating the unemployed (OECD, 2012[34]).57 Despite that reduction in resources per person, the more active stance that had gradually been adopted in the years preceding the crisis remained largely intact and the recessionary increases in long‑term unemployment and labour force withdrawal were lower than would have been predicted, given the severity of the downturn.
A rapid upscaling of re‑employment support is more feasible at the regional level when this expansion is supported by a national effort. For example, the national government in Australia operates Structural Adjustment Programmes that support regions where structural decline in key industries (e.g. autos, steel, textiles and forestry) has resulted in large‑scale displacement (OECD, 2016[28]). The responses to a number of recent natural disasters also illustrate how national governments can support local and regional authorities in responding to a sudden upsurge in the number of displaced workers in a context in which re‑employment services need to be closely co‑ordinated with other government services, such as those related to arranging housing and schooling for families who were evacuated from the affected areas (Venn, 2012[35]). The 2010‑11 earthquakes in Canterbury New Zealand provide an example of the national government ramping up public support for workers displaced by a natural disaster, including measures to save jobs and measures to expand income and re‑employment support for workers whose jobs could not be saved (see Box 4.4). The US response to the large‑scale economic dislocation that followed Hurricane Katrina in 2005 illustrates the additional complexity of mobilising non‑local resources in a Federal system, where each state operates its own unemployment insurance system and re‑employment services. Many workers in New Orleans and other hard‑hit areas in Louisiana were forced to evacuate their home communities, including large numbers who were evacuated to other states. The Louisiana Department of Labor received important help from their counterparts in surrounding states and the Federal government in making it possible for displaced workers to access UI benefits and re‑employment services at evacuation centres.58

4.3.4. What role for targeted re‑employment assistance for displaced workers?

In most OECD countries, displaced workers primarily access public re‑employment services through the general ALMPs that are operated by the PES and do not treat displaced workers as a distinct client group from other jobseekers. However, targeted programmes are sometimes set up for displaced workers or subgroups of displaced workers. Often, these targeted programmes are also operated by the PES as part of their portfolio of services that can be offered to jobseekers, just as they may offer special programmes for unemployed youth, new immigrants, persons with partial disabilities and other groups. In particular, the public early intervention measures that were discussed in Section 4.2 are necessarily organised in this way (e.g. rapid response services for mass layoffs). It is much less evident, however, whether it also makes sense to set up targeted programmes for displaced workers once they have become unemployed and are registered with the public employment service. This is a very heterogeneous group which overlaps considerably, in terms of the re‑employment support that they require, with other job seekers served by the PES. As was discussed above, re‑employment services should be tailored as much as possible to individual needs, but it is not clear how much that goal is furthered by creating targeted services for displaced workers.

A small number of countries, but also the European Union, have set up separate public programmes to provide re‑employment assistance (and sometimes income benefits) to certain subgroups of displaced workers that are considered to require more intensive or somewhat different types of assistance than is provided by general ALMPs. Often, these programmes focus on workers who are adversely affected by increased import competition or were employed in one or a few declining sectors. Trade Adjustment Assistance (TAA) in the United States and the more recently established European Globalisation Adjustment Fund (EGF) in the European Union are the most prominent examples of programmes targeted on trade displaced workers, while Australia has placed a particular emphasis on sectoral adjustment programmes.59 Since the sectors that have been chosen to receive this form of support have been characterised by exposure to intense import competition (e.g. autos, steel, and textiles), the focus of these two types of independent targeted programmes have overlapped to a considerable extent. However, the sectoral programmes in Australia have placed much greater emphasis on measures to enhance sectoral competitiveness and economic redevelopment in local labour markets that are hard‑hit by the sharp erosion of their comparative advantage, whereas EGF and TAA, have emphasised the provision of re‑employment assistance to individual displaced workers.60

Targeted programmes for displaced workers have a number of potential benefits, whether organised as distinct services within the range of ALMPs operated by the PES, or as an entirely separate programme. The main benefit is that targeted programmes can provide a mix of services that is optimised to address the needs of displaced workers, such as the rapid response services offered to a group of workers on notice they will lose their job in a mass layoff. Operating a separate ALMP stream of re‑employment services for displaced workers also makes it easier to develop a group of case workers and counsellors who specialise in working with displaced workers and become expert in the specific issues this group faces. Additional potential benefits of setting up an entirely separate programme for displaced workers are that this approach makes it easier to offer this group more intensive services than are available to unemployed persons generally through the PES, while also being more visible. Increased visibility of public programmes that assist trade displaced workers could be useful for assuaging popular concerns about the adverse impact of globalisation on workers.61

Targeted measures, especially those operated as independent programmes, also have potential drawbacks. These disadvantages have been clearly documented in the case of TAA, which has been the object of a number of careful evaluations since it was established in 1960 (OECD, 2016[27]).62 In particular, running a separate and better resourced programme for a subset of displaced workers, such as trade displaced workers, can create both inefficiencies and inequities:

Inefficiencies can result from the duplication of programmes and administrative processes. In particular, eligibility determination has proven to be a cumbersome and often rather arbitrary process, although it has improved over time. One difficulty is the conceptual and practical difficulty of distinguishing between workers who are displaced because of international trade and those displaced for other reasons, since the extent to which import competition is a causal factor in a particular economic dismissal is both difficult to assess and varies along a continuum from not being a factor to being the only factor. This complexity, together with the concern to effectively control access to an expensive package of government‑financed benefits has led to a burdensome application and review process that has often meant that benefits only became available long after the displacement occurred, reducing their effectiveness.
A second drawback to operating an independent programme for a subset of displaced workers is that it is very likely to create inequities because more intensive support is offered to job losers who qualify for the targeted programme than is offered to other displaced workers (and other jobseekers more generally who face similar barriers to successful adjustment.)

Box 4.4. Assisting workers displaced by the Canterbury earthquakes in 2010‑11

In 2010 and 2011 significant earthquakes struck Christchurch, New Zealand’s second largest city, and its rural hinterland causing rock falls and land damage, widespread building and infrastructure damage and, in the 2011 case, loss of life. The financial cost of the damage, excluding business disruption and clean‑up costs is estimated at 10% of New Zealand’s Gross Domestic Product (APEC, 2013[36]). Following these disasters, population size in Christchurch City fell about 6%, whereas nearby districts in the Canterbury region experienced population increases, partly due to movement out of the city (Reserve Bank of New Zealand, 2016[37]).

Employment in Canterbury initially declined by 5% after the Earthquakes, but has since risen by about 16%, with that rise almost exclusively accounted for by rapid growth in the construction industry, which encountered labour shortages during the rebuilding period.

The national government took a number of initiatives to expand both income and re‑employment support for workers who were displaced by the earthquakes, while also helping local employers to recover.

As concerns income support, people who lost income because they could not get to work or because their workplace closed could get a Civil Defence Payment for loss of livelihood. This pre‑existing programme was paid with an open duration but was relatively modest, providing less than the equivalent of full‑time employment at the minimum wage. A new income benefit programme was set up temporarily for people not qualifying for either the Civil Defence Payment or means‑tested social assistance. The Earthquake Job Loss Cover provided full‑time workers whose employer had closed due to the earthquakes a benefit of NZD 400 per week for a maximum duration of six weeks and a smaller benefit to part‑time workers. About 2% of the workers in the Canterbury region were receiving this benefit in March 2011.

The government also set up a range of active employment services, on top of the existing general system, to assist workers displaced due to this natural disaster. The Earthquake Support Subsidy was a time‑limited employment subsidy that supported small firms in retaining workers during the disaster recovery period. Overall, this subsidy was paid to about 16% of the workers in the greater Canterbury region in March 2011. According to the 2011 Canterbury Employers Survey, 57% of workplaces that received the subsidy said that it “helped a lotˮ in keeping their business going. Two further labour market programmes were also introduced to assist workers whose jobs could not be saved. Jobs for a Local was a wage subsidy programme for jobseekers in the Canterbury region. The jobs created had to be full‑time and permanent, and required the further development of a training plan. The second programme was an extension of the existing Straight to Work programme, where employers were encouraged to train workers to fill labour shortages.

During the rebuilding phase, when worker shortages arose, the Ministry of Social Development (MSD) also introduced a NZD 3k to Christchurch worker mobility subsidy that connected welfare beneficiaries nationwide to the Canterbury labour market by providing a non‑taxable NZD 3 000 payment for applicants who need to relocate to secure sustainable, full‑time employment. Applicants need a confirmed job before relocating and as of June 2015, 1 512 jobseekers were approved for NZD 3k to Christchurch incentive payments.

There do not appear to have been formal evaluations of the effectiveness of any of the earthquake policies. In large part, the lack of evaluations is due to the need for rapid responses and the temporary nature of assistance. Planning and designing evaluations under such crisis circumstances is always unlikely to be a policy priority.

Source: OECD (2017[38]), Back to Work: New Zealand: Improving the Re‑employment Prospects of Displaced Workers, http://dx.doi.org/10.1787/9789264264434-en.

Overall, a cautious but pragmatic approach to the use of general and targeted programmes seems best. Targeted early intervention measures appear to be useful for managing mass layoffs and this type of support should be extended, as much as possible, to workers who are displaced by small‑scale layoffs. Once workers are displaced, the case for setting up targeted programmes is more limited, but should not be dismissed out of hand. Finally, there should be a strong preference to organise targeted measures as options within the portfolio of ALMPs operated by the public employment service, rather than setting them up as a separate programme such as TAA.

4.4. Income support

When quick job‑to‑job transitions are not feasible, income support becomes a key issue for displaced workers. The most important source of public income support for displaced workers is unemployment insurance (UI) and other types of unemployment benefits (UBs). Accordingly, this section focusses on recent policy experience with providing UBs to this group. The biggest challenge for a UB system is to provide income security without undermining work incentives. While this is true for all UB recipients, the forms taken by the tension between benefit coverage and adequacy, on the one hand, and labour supply incentives, on the other, tend to be somewhat different for stable workers who experience a redundancy than for other unemployed jobseekers. While the general principles for designing and operating UI/UB systems also apply to their role in providing income support to displaced workers, there are some important nuances.

The earnings losses associated with unemployment are only one of the sources of the income losses that many displaced workers experience and this has important implications for designing income support for this group, including how extensively UBs should be supplemented by other forms of compensation. Panel A of Table 4.3, identifies four distinct sources of earnings losses for displaced workers and the different types of income support measures that are used to address each type of loss. Panel B then provides an overview of some of the policy design issues that arise for each of the six types of income support measures included in the table. As was discussed in Section 4.2, EPL can be used to require employers to provide compensation for the earnings losses suffered by workers they displace in the form of severance payments. However, any such requirement needs to be used with care because research suggests that they have a high efficiency cost – see OECD (2013[2]), for a survey of the literature. Accordingly, this section does not analyse a policy option of raising mandatory severance levels, but it does consider the implications of severance pay for the operation of public UB systems.

	Types of policy measures
Sources of income loss for displaced workers	Short‑time work (STW) scheme	Unemployment insurance	Means‑tested unemployment and social assistance benefits	Severance payments	Public insurance against unpaid compensation	Wage insurance
Earnings dip prior to displacement	XX		X (Especially, in‑work benefits).
Lost compensation due to insolvency of former employer					XX
Post‑displacement joblessness		XX	X	XX
Loss of earnings once re‑employed (e.g. due to loss of specific human capital or seniority wages)		X (If partial UI benefits paid to workers re‑employed at low earnings).	X (Especially, in‑work benefits).	X		XX

	Types of policy measures
Policy design issues	Short‑time work (STW) scheme	Unemployment insurance	Means-tested unemployment and social assistance benefits	Severance payments	Public insurance against unpaid compensation	Wage insurance
Income targeting criteria	Earnings loss due to lower hours worked.	Earnings loss due to joblessness.	Household income falls below adequacy standard.	None (tied to end of employment relationship).	Compensation left unpaid due to employer insolvency.	Decline in earnings between pre‑ and post‑displacement jobs.
Other targeting criteria	Meet minimum UI eligibility thresholds and employer facing temporarily low demand.	Meet minimum employment/ contribution thresholds; may be means-tested against severance payments.	Asset test(s) common, especially for social assistance.	Usually reflects tenure, sometimes also age or occupation.		Minimum hours worked on new job, sometimes limited to certain groups of displaced workers, such as those aged 50 and older.
Work availability requirement	Sometimes subject to job search or training requirements.	Conditional on work availability and active job search, but nature of activation measures varies considerably.	Conditional on work availability for some beneficiaries, but nature of activation measures varies considerably.	None.	None.	Minimum work hours on new job.
Potential sources of inefficiency and abuse	Impeding efficiency-enhancing mobility by subsidising jobs that are no longer economically viable.	Blunting of labour supply incentives leading to excessive unemployment duration and benefit dependency.	Blunting of labour supply incentives leading to excessive unemployment duration and benefit dependency.	Blunting of incentives for efficiency – enhancing mobility (pre‑displacement) and labour supply (post‑displacement).	Blunting of incentives for employers to pre‑fund deferred compensation (and for workers/unions to insist that they do).	Blunting incentives to find a new job that pays as well as the lost job or to work full time.
Prominence of displaced workers among all beneficiaries	In principle, all beneficiaries risk displacement, but some deadweight is likely (i.e. subsidies are paid for jobs that would have been preserved in any case).	Displaced workers are one of the main groups targeted.	Persons facing longer-term disadvantages are the main target group.	Displaced workers are the main target group.	A subset of displaced workers is the target group.	A subset of displaced workers is the target group.
How widely is this policy measure used in OECD countries?	24 out of 34 OECD countries, but take-up is low in many cases.	Widespread, but not universal (e.g. Australia and New Zealand have means‑tested social assistance with benefit levels that reflect family income needs, rather than the level of past earnings).	Widespread, but coverage and generosity vary considerably.	Mandatory severance in 22 out of 34 OECD countries. Collective bargaining and firm human resource policies provide for severance (or additional severance) for some workers.	Widespread, but not universal	Small programmes in only a few countries (e.g. France, Korea and the United States), but gradual benefit phase‑out for UI beneficiaries accepting low‑paying jobs has a similar effect and is more widespread.
Other policy issues (highly selective)	Should STW be combined with mandatory job search or training?	Should UI benefit eligibility be delayed until severance payments have been spent down?	Should there be time limits or mandatory workfare?	Should legally mandated severance take the form of portable retirement savings accounts, so as to avoid penalising voluntary labour mobility?	How should this insurance relate to bankruptcy law, where workers are only one of multiple creditors?	Affordability of a broad wage insurance scheme remains to be demonstrated.

Note: In Panel A, XX and X denote, respectively, a major and secondary policy goal of the indicated policy. UI: Unemployment insurance.

As was already emphasised in Section 4.1, the two main sources of earnings losses for displaced workers are those associated with the period of joblessness and zero earnings that follows most layoffs and the longer‑term losses due to re‑employment at a lower level of earnings. While these losses and the ways in which they can be compensated will be the main focus of this section, it is useful also to briefly discuss two additional sources of earnings losses for displaced workers, namely, a tendency for earnings to decline in the period immediately preceding displacement (e.g. as hours of work are reduced in a struggling firm) and the risk that a firm entering bankruptcy will fail to pay their employees all of the compensation to which they are entitled. Both short‑time working schemes and in‑work benefits provide some compensation for a pre‑displacement dip in earnings.63 As regards unpaid compensation, some OECD countries have established public insurance schemes to compensate such losses, such as the Wage Earner Protection Plan (WEPP) that the Canadian government established in 2008 (OECD, 2015[13]).64

4.4.1. Unemployment benefits

Benefit adequacy

A generous unemployment benefit system represents one of the most straightforward policy approaches for reducing the costs borne by displaced workers. However, governments need to carefully balance the direct benefits for workers, who are better compensated for their earnings losses and can therefore smooth their consumption over time as well as have sufficient resources to look for a job that matches their skills and expectations, against the disincentive effects on individual job search effort (moral hazard effect) as well as possible aggregate effects on labour supply, labour demand and the government budget – see e.g. Tatsiramos and van Ours (2014[39]); Schmieder and von Wachter (2016[40]); Nekoei and Weber (2017[41]). Whether or not the moral hazard effect of UI benefits is particularly large for displaced workers, as compared to other UI recipients, remains an open question. Nevertheless, the evidence suggests that there is an effect and governments need to find a middle ground between generously compensating displaced workers for their lost earnings and encouraging rapid re‑employment.

Consistent with the benefit entitlement rules for UI programmes, displaced workers with a stable work history tend to have relatively high benefit levels, as compared to job losers with less continuous employment histories. Nonetheless, Figure 4.8 shows that the level of income support that results is much higher in some OECD countries than in others. Focusing on the nine countries that participated in the Back to Work reviews, the figure shows that average net replacement rates (NRRs) during the first year of unemployment vary from 19% in the United States to 75% in Denmark. This gap reflects both the higher initial benefit level in Denmark (net replacement rates of 75% in Denmark versus 50% in the United States) and the longer maximum duration of benefit payments (24 versus 4.6 months). Since a considerable share of displaced workers experience long spells of unemployment, the share of earnings losses due to joblessness that is compensated by UBs will be significantly lower in countries where the maximum period of benefit receipt is relatively short.65 For example, one in four displaced workers in the United States in 2014 had exhausted their UI entitlement – OECD (2016[27]). While many displaced workers do not qualify for unemployment benefits in Australia and New Zealand, due to means testing (see below), and the initial NRRs are quite low for qualifying job losers in these countries, the absence of a time‑limit on the receipt of these benefits means that the average NRRs over 5 years are substantially above the OECD average.

Given the high level of concern about the hardship experienced by displaced workers and the relatively high risk they experience long spells of unemployment, it is natural to ask whether UB rules should treat displaced workers more generously than other unemployed jobseekers. To a limited extent this is quite common. For example, eligibility rules often either deny benefits to certain groups, such as workers who voluntarily left their job, self‑employed or apprentices whose contract is not renewed, or impose an additional waiting period on such applicants before they are entitled to begin receiving benefits. However, the Employment Insurance (EI) programme in Japan provides an example of a much more ambitious approach to providing greater income support to formerly stable workers who are displaced and then are slow to become re‑employed than is available to other jobseekers (OECD, 2015[23]). The maximum duration of EI payment is significantly longer for specific qualified recipients, a category which mostly applies to displaced workers, than it is for ordinary unemployed, a category which includes most people voluntarily quitting their job or whose temporary job ended. The maximum EI payment duration also increases quite strongly with age and job tenure for specific qualified recipients (e.g. from 90 days for a displaced worker aged 44 years or younger with 1 to 5 years of job tenure, at the low end, up to 330 days for a displaced worker aged 45‑59 with 20 or more years of tenure). Such an approach probably better aligns the level of income support with the risk of long‑term unemployment, but should only be pursued if combined with more intensive activation of the groups eligible for longer duration benefit payments.

Coverage rates

How effectively UB systems compensate for the earnings losses is also influenced by the share of displaced workers who qualify to receive these benefits, that is, the effective coverage rate for this group.66 That share appears to be quite high in seven of the review countries where the first tier UB system is organised as unemployment insurance (see Chapter 5). Nonetheless, coverage gaps are of concern for certain workforce groups. For example, effective UI coverage is relatively low among non‑regular workers who are displaced in Korea and Japan. However, coverage rates are on an upward trend in Korea (as the UI system matures) and Japan experimented successfully with a temporary extension of UI eligibility to more low tenure workers during the economic crisis, possibly setting the stage for permanent measures (OECD, 2013[42]; 2015[23]). There has been a downward trend in UI coverage rates in Denmark and Sweden, where enrolment is voluntary, but only Sweden has a basic public unemployment benefit that is available to job losers who chose not to enrol in a UI fund (OECD, 2015[26]; 2016[43]). Declining coverage in Denmark is also highly concentrated in the bottom three income deciles causing lower‑income displaced workers to be particularly unlikely to receive income benefits, even when they experience a lengthy spell of unemployment.67 By far the largest gaps in UB coverage are found in Australia and New Zealand, where the first tier UB programme is designed as a safety net of last resort that provides a flat‑rate payment to families whose income and liquid assets are below the minimum adequacy standards set by the government (OECD, 2016[28]; 2017[38]). This results in relatively few displaced workers qualifying for public income support following displacement, at least initially, although more become eligible eventually if they remain unemployed for an extended period and their spouse has little or no earnings. For example, only about one‑third of the stock of non‑employed displaced workers reported welfare benefit receipt in 2015 in New Zealand.

Interaction with other sources of income support

In assessing the adequacy of the income support that displaced workers receive from UB programmes, it is important also to take account of both the severance awards received by many displaced workers and other public programmes, especially the social assistance programmes that act as a backstop to the first tier UB scheme. Severance awards are quite widespread in some OECD countries and can be quite high. For example, long‑tenure regular workers in large corporations in Japan tend to accumulate severance entitlements that exceed their maximum cumulative UI benefit entitlement (OECD, 2015[23]). However, it appears that the displaced workers with the greatest UB entitlements also tend to receive the most severance, suggesting a limited role for severance payments in plugging the most worrisome gaps in UB adequacy.68 Another indication that severance awards and last‑resort social benefits are of limited effectiveness in avoiding large uncompensated earnings losses following displacement is that the poverty risk for displaced workers appears to be quite high in some of the countries studied. For example, in the United States, two in three families with a displaced worker fall into poverty for some time (OECD, 2016[27]).

Figure 4.8. Unemployment benefit schemes are a key source of income support

The Back to Work country reviews highlighted an issue that has not received a lot of attention, namely, how unemployment benefit entitlements and severance payments should be co‑ordinated. Table 4.4 summarises legal entitlements to severance pay in the reviewed countries (and several other OECD countries), as well as any rules about interactions between severance pay and UB payments. Quite often, the receipt of severance pay delays UB entitlement with this effect being particularly strong in Australia, Canada, Finland and Sweden. In Denmark and Sweden, the size of the severance payment declines as the UI benefit increases. These various off‑sets may reflect judgements about the overall adequacy of the income support provided by the combination of these two types of payments. While the question how best to co‑ordinate UB and severance payments is understudied and it would be premature to designate best practice principles, these examples raise several issues that merit further attention:

	Legal basis and eligibility conditions	Minimum amount set by statutory law	Collective bargaining coverage rate^a	Interaction with UI entitlement
Australia	Federal statutory law and provisions in collective agreements.	Min: tenure < 1 year = 0, tenure ≥ 1 year and < 2 year = 4 weeks. Max: tenure ≥ 9 years and < 10 years = 16 weeks. Tenure ≥ 10 years = 12 weeks.	59% (2016).	Waiting period for UI benefits is increased by number of (wage) days received in severance pay.
Austria^b	Statutory law: Access to individual accounts only if tenure over 3 years. Otherwise account carried over to next employer.	Amount depends on the capital accrued in the fund, investment income earned and capital guaranteed.	98% (2016).	None.
Canada	Federal statutory law, Provincial law, and provisions in collective agreements.	Min (employees covered by federal law): tenure < 1 year = 0, tenure ≥ 1 year and < 3 years = 5 days after which tenure ≥ 3 years = 2 days for each year of tenure. Min (Ontario): tenure < 1 year = 0, Max (Ontario): Tenure ≥ 26 years = 26 weeks if the firm has a payroll of CAD 2.5 million or more. Other jurisdictions: no legislated severance pay.	28% (2016).	Waiting period for UI benefits is increased by number of (wage) days received in severance pay.
Denmark^c	Statutory requirement for white collar workers and collective agreements ofor blue collar workers.	White collars: Min: Tenure < 12 years = 0, Tenure ≥ 12 years and < 15 years = 1 month. Max: Tenure ≥ 18 years = 3 months. Blue collars: The monthly amount of severance pay is calculated as follows: 85% of monthly salary minus the monthly unemployment benefit, and is payable for: 1 month > 3‑year tenure; 2 months > 6‑year tenure; 3 months > 8‑year tenure.	84% for all workers (2015).	For blue‑collar workers, the amount of severance pay is reduced by the amount of UI benefits. Indeed, since initial replacement rates are most often above 85% for low‑paid workers, severance pay is rarely paid to blue collars.
Finland	No legal requirement. Provisions in collective agreements.	n.a.	89% (2015).	Waiting period for UI benefits is increased by the number of (wage) days received in severance pay.
France	Statutory law and provisions in collective agreements.	Min: tenure < 8 months: 0, tenure ≥ 8 months and < 10 years: 0.25 months per year of service, tenure ≥ 10 years: 1/3 month per year of service	99% (2014).	Waiting period for UI benefits is increased if severance pay exceeds legal minima, by a duration in days corresponding to the extra‑amount in severance pay divided by 90 (total waiting period capped at 75 days).
Germany	Statutory law: an employee working in a firm with at least ten employees who is dismissed on the basis of compelling operational reason is entitled to severance pay if offered by the employer and the workers renounces to go to court.	Half a month’s pay per year of tenure (if offered by the employer).	56% (2016).	Waiting period for UI benefits is increased by a fraction of the number of (wage) days of severance, where the fraction varies with age and job tenure and the delay is capped at 1 year.
Japan	No legal requirement. Provisions in collective agreements.	n.a.	17% (2016).	None.
Korea	No legal requirement. Provisions in collective agreements.	n.a.	12% (2015).	3‑month delay in UI benefits if severance pay is KRW 100 million or more.
New Zealand^d	No statutory requirements in the Employment Relations Act. Except under some circumstances for a very small group of “vulnerableˮ workers.	Paid if explicitly negotiated and included in individual or collective employment agreements.	20% (2016).	No interaction, except one week longer benefit stand-down (i.e. two weeks) if redundancy pay pushes prior annual income over the average annual income.
Sweden	No legal requirement. General provisions established in collective agreements respectively for white collars aged over 40 and for blue collars aged over 40 and with 50 months of employment over the last 5 years.	White collars: Complements UI at a max of 70% of previous wage for a period of 6‑18 months depending on age. Blue collars are entitled to a lump sum increasing with age.	90% (2015).	Waiting period for UI benefits is increased by number of (wage) days received in severance pay. Severance pay amount declines with UI benefit level for white‑collar workers.
United States	No legal requirement. Provisions in collective agreements.	n.a.	12% (2016).	Increased waiting period for UI benefits or reduction in the benefit amount depending on the state.

Note: UI: Unemployment insurance; n.a.: Not applicable.

a. The collective bargaining coverage rate provides an indication of the proportion of the workforce potentially covered under these agreements and therefore likely to receive higher severance packages than the legislated ones.

b. Austria: Conditions refer to workers with contracts concluded after January 2003.

c. Denmark: Conditions are regulated by collective agreements per sector for blue‑collar workers and by regulation for white‑collar workers.

d. New Zealand: In case of restructuring, defined as outsourcing, the employee has the right to ask for transfer to the contractor. If refused, the worker can negotiate redundancy arrangements.

Source: For statutory severance pay: Décret n° 2017-1398 du 25 septembre 2017, https://www.legifrance.gouv.fr/eli/decret/2017/9/25/2017-1398/jo/texte, for France, and OECD Employment Protection Database, 2013 update, www.oecd.org/employment/protection, for other countries; OECD/ICTWSS Database, https://stats.oecd.org/Index.aspx?DataSetCode=CBC, for adjusted bargaining coverage rates; and country responses to the “OECD Questionnaire on Eligibility Criteria for Unemployment Benefits and Interventions in the Unemployment Spellˮ for interaction with UI entitlement (rules as of June 2014).

As was discussed in Section 4.3, delaying eligibility for UBs until severance pay has been spent down (as appears to be the logic of the offset rules used in Australia, Canada, Finland and Sweden), has important implications for the provision of re‑employment support to displaced workers. In particular, a strategy is then needed to connect displaced workers to re‑employment services even before they become eligible for UB payments
If UB and severance payments are considered to represent alternative sources of compensation for the earnings losses experienced by displaced workers, then it is natural to think of them as being close substitutes and to means test eligibility for one of these payments based on how much of the other is received. However, it might be more appropriate to think of UB benefits as providing partial insurance against the earnings losses due to post‑displacement unemployment, whereas severance payments provide insurance against the loss of earnings due to lower wages upon re‑employment. Parsons (forthcoming[44]) provides a theoretical argument supporting the idea that a well‑designed combination of UI and severance pay could represent an efficient form of “job displacement insurance”. In that package, UI provides insurance against the unemployment risk while severance pay provides insurance against wage loss. From this perspective, it probably does not make sense to think of these two types of insurance as being close substitutes and to means test one against the other.
Considered as insurance for the wage loss associated with displacement, severance pay takes the form of a scheduled (i.e. lump sum) benefit, the payment of which is triggered by displacement while the amount paid is independent of the size of the actual wage losses. An alternative design for providing wage insurance is to structure it as a public social insurance programme where eligibility to receive a benefit is conditional on re‑employment at a lower wage and the size of the payment depends on ex‑post wage losses.

4.4.2. Is there a role for wage insurance?

A major unresolved issue related to the provision of income support to displaced workers is whether and how to compensate for the part of earnings losses that sometimes persists long after they have become re‑employed, because they can no longer command as high of a wage as they earned on the lost job. Particularly for long‑tenure blue‑collar workers, this can represent the largest part of total earnings losses in the long‑run. Wage insurance (WI) is sometimes proposed as a supplement to unemployment insurance which cushions this second type of earnings loss following displacement. Similarly, to unemployment benefits, which offset a portion of the earnings losses due to post‑displacement unemployment, WI pays displaced workers who accept new jobs at lower wages an earnings supplement that replaces a fraction of the difference between earnings on the old and the new job. Often, it is proposed that this supplement would be limited in duration (maybe one or two years) or limited to certain groups who are particularly at risk of experiencing a permanent loss of earnings capacity, such as older and long‑tenured displaced workers, and/or workers who become re‑employed within a certain period (e.g. within six months after displacement). The experience to date with WI is quite limited, but proposals to implement it on a larger scale have been a recurrent feature of employment policy discussions in North America the past several decades (OECD, 2015[13]; 2016[27]).69

Proponents of WI argue that it can provide a more equitable sharing of the gains from economic restructuring by reducing the adjustment costs faced by those who are hurt the most. It is also argued that WI would improve incentives for speedy re‑employment, as unemployment benefits become less attractive relative to accepting a new job. But WI also comes with potential problems. First, the cost could be high unless the earnings supplements are tightly targeted. Tight targeting, however, would be likely to create inequities between displaced workers re‑employed at lower wages who receive the WI benefit and similar workers who do not. To the extent WI speeds up re‑employment, shorter unemployment spells could come at the cost of shifting workers into low‑quality jobs with low wages and poor prospects for training and wage growth. Finally, there may be a risk of employers being able to offer lower wages than they would in the absence of such schemes.

Evaluations of two small wage insurance schemes in North America suggests that WI is an effective instrument for reducing the decline in the net incomes of displaced workers who become re‑employed at a lower wage, but does not have a significant impact in speeding re‑employment or affecting the post‑displacement wage (Bloom et al., 2001[45]; Wandner, 2016[46]). While it seems premature to implement a large‑scale wage insurance scheme in the absence of a clearly demonstrated working model, further pilot studies of WI schemes would be of considerable value given the high level of concern about the impact of displacement on worker well‑being. It would also be useful for researchers to assess the comparative advantages and disadvantages of WI as compared to alternative measures that also be used to compensate displaced workers who become re‑employed at a lower wage, such as a gradual phasing out of UI benefits as re‑employment earnings rise, severance pay and general in‑work benefit schemes.

4.5. Concluding remarks

Reconciling economic dynamism with employment and income security for workers is an important policy challenge and labour market programmes have a central role to play in meeting that challenge. This chapter has analysed how best labour market programmes can play that role by summarising the main lessons from the OECD’s recent Back to Work reviews in nine countries. It highlights a number of effective practices that are already in place in some OECD countries, as well as a number of areas where improvement is needed. The latter include reducing both the sometimes large discrepancies in the assistance provided to different groups of displaced workers facing similar difficulties in reintegrating into the labour market as well as the sometimes long delays in connecting displaced workers with the re‑employment services they require.

The chapter confirms that the starting point for improving the re‑employment prospects and income security of displaced workers is to make further progress at developing a system of well‑designed and adequately‑resourced active and passive labour market programmes that implement an effective national activation strategy while also providing an adequate level of income security. However, the general principles of good labour market practice need to be applied in a way that addresses the particular situation of displaced workers, including both the particular barriers to successful re‑employment that they face and their particular advantages in searching for a new job. From the perspective of activation policy, two of the most important differences between displaced workers and other jobseekers are the greater scope for proactive measures, beginning during the notice period before the layoff occurs, and the large contribution that employers can make to fostering successful mobility for workers they displace, ideally in close collaboration with trade unions and labour market authorities.

The chapter identifies a number of avenues for improving the assistance that labour market programmes provide to displaced workers, but also leaves key questions unanswered. One such question concerns spending priorities. Many of the strategies identified here for improving re‑employment services and income‑support for displaced workers would imply higher spending. In the context of tight budgetary constraints, expanding the resources devoted to assisting displaced workers could lead to a reduction of the resources available to help other groups (e.g. the long‑term unemployed, welfare benefit recipients and people with partial disabilities), who have been a major focus of activation policy in recent years. Spending priorities will need to be set in light of specific national conditions, but the currently high level of concern about mass layoffs suggests it is timely to consider redoubling efforts to assist displaced workers. Even if it were not deemed appropriate to increase spending on assistance for this group, the chapter may provide useful guidance for using existing resources more effectively.

References

[18] Albrecht, J. and S. Vroman (1999), “Unemployment Compensation Finance and Efficiency Wages”, Journal of Labor Economics, Vol. 17/1, pp. 141-167, http://dx.doi.org/10.1086/209916.

[62] Andersson, J. (2017), Insurances against job loss and disability: Private and public interventions and their effects on job search and labor supply, Uppsala University, http://www.nek.uu.se/digitalAssets/244/c_244210-l_3-k_josefine-andersson.pdf (accessed on 26 April 2018).

[8] Andrews, D. and A. Saia (2017), “Coping with creative destruction: Reducing the costs of firm exit”, OECD Economics Department Working Papers, No. 1353, OECD Publishing, Paris, http://dx.doi.org/10.1787/bbb44644-en.

[36] APEC (2013), Building Natural Disaster Response Capacity Sound Workforce Strategies for Recovery and Reconstruct, https://www.apec.org/Publications/2014/02/Building-Natural-Disaster-Response-Capacity--Sound-Workforce-Strategies-for-Recovery-and-Reconstruct (accessed on 21 February 2018).

[15] Arni, P. (2012), Conseil et coaching intensifs pour demandeurs d'emploi âgés : une voie pour améliorer leurs chances sur le marché du travail ?, Université de Lausanne, Lausanne, https://works.bepress.com/patrick_arni/3/.

[61] Arni, P., R. Lalive and J. Van Ours (2013), “How effective are unemployment benefit sanctions? Looking beyond unemployment exit”, Journal of Applied Econometrics, Vol. 28/7, pp. 1153-1178, http://dx.doi.org/10.1002/jae.2289.

[3] Autor, D., D. Dorn and G. Hanson (2016), “The China Shock: Learning from Labor-Market Adjustment to Large Changes in Trade”, Annual Review of Economics, Vol. 8, pp. 205-240, http://dx.doi.org/10.1146/annurev-economics-080315-015041.

[60] Barnow, B. and J. Smith (2015), “Employment and Training Programs”, in Robert Moffitt (ed.), Economics of Means-Tested Transfer Programs in the United States, Volume 2, National Bureau of Economic Research, Inc, https://econpapers.repec.org/bookchap/nbrnberch/13490.htm (accessed on 19 February 2018).

[9] Bassanini, A. and E. Caroli (2015), “Is Work Bad for Health? The Role of Constraint versus Choice”, Annals of Economics and Statistics 119/120, pp. 13-37, http://dx.doi.org/10.15609/annaeconstat2009.119-120.13.

[59] Bassanini, A. and R. Duval (2006), “Employment Patterns in OECD Countries: Reassessing the Role of Policies and Institutions”, OECD Social, Employment and Migration Working Papers, No. 35, OECD Publishing, Paris, http://dx.doi.org/10.1787/702031136412.

[17] Blanchard, O. and J. Tirole (2008), “The Joint Design of Unemployment Insurance and Employment Protection: A First Pass”, Journal of the European Economic Association, Vol. 6/1, pp. 45-77, http://dx.doi.org/10.1162/JEEA.2008.6.1.45.

[45] Bloom, H. et al. (2001), “Testing a Financial Incentive to Promote Re-employment among Displaced Workers: The Canadian Earnings Supplement Project (ESP)”, Journal of Policy Analysis and Management, Vol. 20/3, pp. 505-523, http://dx.doi.org/10.1002/pam.1005.

[7] BLS (2016), Displaced Workers Summary, Economic News Release, Bureau of Labor Statistics, Washington, D.C., https://www.bls.gov/news.release/disp.nr0.htm.

[21] Boeri, T. and H. Bruecker (2011), “Short-time work benefits revisited: some lessons from the Great Recession”, Economic Policy, 10.1111/j.1468-0327.2011.271.x, pp. 697-765, http://dx.doi.org/10.1111/j.1468-0327.2011.271.x.

[19] Cahuc, P. and F. Malherbet (2004), “Unemployment compensation finance and labor market rigidity”, Journal of Public Economics, Vol. 88/3-4, pp. 481-501, http://dx.doi.org/10.1016/S0047-2727(03)00018-5.

[20] Cahuc, P. and S. Nevoux (2017), “Inefficient Short-Time Work”, IZA Discussion Papers, No. 11010, IZA, http://dx.doi.org/www.iza.org.

[11] Card, D., J. Kluve and A. Weber (2015), “What Works? A Meta Analysis of Recent Active Labor Market Program Evaluations”, IZA Discussion Papers, No. 9236, IZA, http://ftp.iza.org/dp9236.pdf (accessed on 19 February 2018), http://ftp.iza.org/dp9236.pdf.

[14] DARES analyses (2017), Le contrat de sécurisation professionnelle favorise-t-il la reprise d'emploi des licenciés économiques qui y adhèrent ?, DARES, http://dares.travail-emploi.gouv.fr/IMG/pdf/2017-020.pdf.

[58] Davis, S. and T. von Wachter (2011), “Recessions and the Costs of Job Loss”, Brookings Papers on Economic Activity, Vol. Fall, https://www.brookings.edu/wp-content/uploads/2011/09/2011b_bpea_davis.pdf (accessed on 19 February 2018), pp. 1-72.

[57] Deelen, A., M. de Graaf-Zijl and W. van den Berge (2018), “Labour market effects of job displacement for prime-age and older workers”, IZA Journal of Labor Economics, Vol. 7/3, http://dx.doi.org/10.1186/s40172-018-0063-x.

[56] Department of Jobs and Small Business (2018), Stronger Transitions, https://www.jobs.gov.au/stronger-transitions (accessed on 28 April 2018).

[24] Diedrich, A. and O. Bergström (2006), “The Job Security Councils in Sweden”, http://imit.se/wp-content/uploads/2016/02/2007_145.pdf.

[31] Duell, N. et al. (2010), “Activation Policies in Switzerland”, OECD Social, Employment and Migration Working Papers, No. 112, OECD Publishing, Paris, http://dx.doi.org/10.1787/5km4hd7r28f6-en.

[25] European Commission (2010), 27 National Seminars on anticipating and managing restructuring, EU Synthesis Report, http://www.employment-studies.co.uk/resource/27-national-seminars-anticipating-and-managing-restructuring-arenas.

[55] Farber, H. (2017), “Employment, Hours, and Earnings Consequences of Job Loss: US Evidence from the Displaced Workers Survey”, Journal of Labor Economics, doi: 10.1086/692353, pp. S235-S272, http://dx.doi.org/10.1086/692353.

[54] Farber, H. (2004), “Job loss in the United States, 1981–2001”, http://dx.doi.org/10.1016/S0147-9121(04)23003-5.

[16] Filges, T. and A. Hansen (2017), “The threat effect of active labor market programs: a systematic review”, Journal of Economic Surveys, Vol. 31/1, pp. 58-78, http://dx.doi.org/10.1111/joes.12134.

[5] Fujii, M. and R. Kambayashi (2014), “Long-term effects of job displacement in Japan: A conservative estimate using the Japanese Longitudinal Survey on Employment and Fertility (LOSEF)”, Hermes-IR Technical Report, No. 2014-10, Hitotsubashi University, https://hermes-ir.lib.hit-u.ac.jp/rs/bitstream/10086/26917/1/DP634.pdf.

[53] Gibbons, R. and L. Katz (1991), “Layoffs and Lemons”, Journal of Labor Economics, Vol. 9/4, http://www.jstor.org/stable/2535075, pp. 351-380.

[52] Goldstein, A. (2017), Janesville : an american story., Simon & Schuster, New York, NY, http://www.simonandschuster.com/books/Janesville/Amy-Goldstein/9781501102264 (accessed on 15 January 2018).

[51] Immervoll, H. (2012), “Reforming the Benefit System to 'Make Work Pay': Options and Priorities in a Weak Labour Market”, IZA Policy Papers, No. 50, IZA, Bonn.

[65] Immervoll, H. and C. Knotz (forthcoming), “How demanding are activation requirements for jobseekers? New evidence on activity-related eligibility criteria for unemployment and social assistance benefits”, OECD Social, Employment and Migration Working Papers, OECD Publishing, Paris.

[30] Immervoll, H. and S. Scarpetta (2012), “Activation and employment support policies in OECD countries. An overview of current approaches”, IZA Journal of Labor Policy, Vol. 1/1, p. 9, http://dx.doi.org/10.1186/2193-9004-1-9.

[12] Kluve, J. (2010), “The effectiveness of European active labor market programs”, Labour Economics, Vol. 17/6, pp. 904-918, http://dx.doi.org/10.1016/J.LABECO.2010.02.004.

[32] Mosley, H. (2010), Reform of placement services - Perr Review on 'Systemic preventive integration approach (Support) for jobseekers and unemployed, EC Mutual Learning Programme, http://ec.europa.eu/social/BlobServlet?docId=10712&langId=en.

[50] Nafilyan, V. (2016), “Lost and found?: The cost of job loss in France”, OECD Social, Employment and Migration Working Papers, No. 194, OECD Publishing, Paris, http://dx.doi.org/10.1787/5jlsk8tzll42-en.

[41] Nekoei, A. and A. Weber (2017), “Does extending unemployment benefits improve job quality?”, American Economic Review, http://dx.doi.org/10.1257/aer.20150528.

[49] Nord, S. and Y. Ting (1991), “The impact of advance notice of plant closings on earnings and the probability of unemployment”, Industrial and Labor Relations Review, Vol. 44/4, pp. 681-691.

[38] OECD (2017), Back to Work: New Zealand: Improving the Re-employment Prospects of Displaced Workers, Back to Work, OECD Publishing, Paris, http://dx.doi.org/10.1787/9789264264434-en.

[10] OECD (2017), OECD Employment Outlook 2017, OECD Publishing, Paris, http://dx.doi.org/10.1787/empl_outlook-2017-en.

[28] OECD (2016), Back to Work: Australia: Improving the Re-employment Prospects of Displaced Workers, Back to Work, OECD Publishing, Paris, http://dx.doi.org/10.1787/9789264253476-en.

[43] OECD (2016), Back to Work: Denmark: Improving the Re-employment Prospects of Displaced Workers, Back to Work, OECD Publishing, Paris, http://dx.doi.org/10.1787/9789264267503-en.

[22] OECD (2016), Back to Work: Finland: Improving the Re-employment Prospects of Displaced Workers, Back to Work, OECD Publishing, Paris, http://dx.doi.org/10.1787/9789264264717-en.

[27] OECD (2016), Back to Work: United States: Improving the Re-employment Prospects of Displaced Workers, Back to Work, OECD Publishing, Paris, http://dx.doi.org/10.1787/9789264266513-en.

[13] OECD (2015), Back to Work: Canada: Improving the Re-employment Prospects of Displaced Workers, Back to Work, OECD Publishing, Paris, http://dx.doi.org/10.1787/9789264233454-en.

[23] OECD (2015), Back to Work: Japan: Improving the Re-employment Prospects of Displaced Workers, Back to Work, OECD Publishing, Paris, http://dx.doi.org/10.1787/9789264227200-en.

[26] OECD (2015), Back to Work: Sweden: Improving the Re-employment Prospects of Displaced Workers, Back to Work, OECD Publishing, Paris, http://dx.doi.org/10.1787/9789264246812-en.

[63] OECD (2015), Fit Mind, Fit Job: From Evidence to Practice in Mental Health and Work, Mental Health and Work, OECD Publishing, Paris, http://dx.doi.org/10.1787/9789264228283-en.

[29] OECD (2015), OECD Employment Outlook 2015, OECD Publishing, Paris, http://dx.doi.org/10.1787/empl_outlook-2015-en.

[64] OECD (2014), Ageing and Employment Policies: France 2014: Working Better with Age, Ageing and Employment Policies, OECD Publishing, Paris, http://dx.doi.org/10.1787/9789264207523-en.

[4] OECD (2013), Back to Work: Re-employment, Earnings and Skill Use after Job Displacement, OECD, Paris, http://www.oecd.org/employment/emp/Backtowork-report.pdf.

[42] OECD (2013), Korea: Improving the Re-employment Prospects of Displaced Workers, Back to Work, OECD Publishing, Paris, http://dx.doi.org/10.1787/9789264189225-en.

[2] OECD (2013), OECD Employment Outlook 2013, OECD Publishing, Paris, http://dx.doi.org/10.1787/empl_outlook-2013-en.

[34] OECD (2012), OECD Employment Outlook 2012, OECD Publishing, Paris, http://dx.doi.org/10.1787/empl_outlook-2012-en.

[33] OECD (2010), OECD Employment Outlook 2010: Moving beyond the Jobs Crisis, OECD Publishing, Paris, http://dx.doi.org/10.1787/empl_outlook-2010-en.

[1] OECD (2009), OECD Employment Outlook 2009: Tackling the Jobs Crisis, OECD Publishing, Paris, http://dx.doi.org/10.1787/empl_outlook-2009-en.

[6] OECD (2005), OECD Employment Outlook 2005, OECD Publishing, Paris, http://dx.doi.org/10.1787/empl_outlook-2005-en.

[44] Parsons, D. (forthcoming), “The simple analytics of job displacement insurance”, Journal of Risk and Insurance, http://dx.doi.org/10.1111/jori.12216.

[37] Reserve Bank of New Zealand (2016), The Canterbury rebuild five years on from the Christchurch earthquake, Bulletin, Wellington, http://www.rbnz.govt.nz/email-updates.

[40] Schmieder, J., T. von Wachter and S. Bender (2016), “The Effect of Unemployment Benefits and Nonemployment Durations on Wages”, American Economic Review, Vol. 106/3, pp. 739-777, http://dx.doi.org/10.1257/aer.20141566.

[48] SSI Task Force (2017), SSI Task Force Legacy Report, http://www.redcar-cleveland.gov.uk/taskforce.nsf/c7ec965d5c9e6c0b80257f930030066e/$File/SSI%20Task%20Force%20Legacy%20Report%20Two%20Years.pdf (accessed on 28 April 2018).

[47] Swaim, P. and M. Podgursky (1990), “Advance Notice and Job Search: The Value of an Early Start”, The Journal of Human Resources, Vol. 25/2, pp. 147-178, http://dx.doi.org/10.2307/145752.

[39] Tatsiramos, K. and J. van Ours (2014), “Labor Market Effects of Unemployment Insurance Design”, Journal of Economic Surveys, Vol. 28/2, pp. 284-311, http://dx.doi.org/10.1111/joes.12005.

[35] Venn, D. (2012), “Helping Displaced Workers Back Into Jobs After a Natural Disaster: Recent Experiences in OECD Countries”, OECD Social, Employment and Migration Working Papers, No. 142, OECD Publishing, Paris, http://dx.doi.org/10.1787/5k8zk8pn2542-en.

[46] Wandner, S. (2016), “Wage Insurance as a Policy Option in the United States”, Upjohn Institute Working Papers, No. 16-250, Upjohn Institute for Employment Research, Kalamazoo, MI, http://dx.doi.org/10.17848/wp16-250.

Notes

← 1. Other terms for this group include redundant, retrenched and laid-off workers. Similarly, job displacements are often referred to as redundancies, retrenchments, layoffs and economic dismissals.

← 2. ^.The ageing of the workforce also increases the risk that rapid structural change in the labour market inflicts large costs on displaced workers, since labour market mobility is particularly difficult for older workers (OECD, 2014[64]).

← 3. The potential fragility of this political support is underlined by the recent successful exploitation of widespread concerns about job losses by populist political movements in a number of OECD countries.

← 4. See OECD (2013[42]) for Korea, OECD (2015[23]) for Japan, OECD (2015[13]) for Canada, OECD (2015[26]) for Sweden, OECD (2016[28]) for Australia, OECD (2016[43]) for Denmark, OECD (2016[22]) for Finland, OECD (2016[27]) for the United States, and OECD (OECD, 2017[38]) for New Zealand. The OECD Secretariat is grateful to the national authorities and many other stakeholders in the nine countries that participated in the OECD Back to Work reviews of policies to assist displaced workers back into suitable jobs. The analysis underlying this chapter could not have been successfully conducted without their generous support.

← 5. The nine countries reviewed are quite diverse yet were found to be grappling with very similar issues in their efforts to support displaced workers. This suggests that these reviews are likely to be informative on the main policy issues that need to be addressed in other OECD countries. It should be noted, however, that no Latin American or non‑Nordic European countries participated in the reviews and it is possible that the chapter’s analysis fails to address specific aspects of the policy challenges facing such countries.

← 6. A recent example of this genre is Amy Goldstein’s book about the 2008 closing of a large General Motors plant in Janesville, Wisconsin (Goldstein, 2017[52]).

← 7. For the purpose of this analysis, the term displaced worker refers to workers involuntarily separated from their job due to economic or technological reasons, such as layoffs related to a recession or structural economic change. Two distinct approaches were used to differentiate job displacement from other types of separations, such as voluntary quits, depending on the underlying data source in each country: i) self‑defined displacement – when household survey data is used, the worker’s assessment of the reason for the separation is used to identify displacements; and ii) firm‑identified displacement – when linked employer‑employee longitudinal data (usually from administrative sources) is used, job displacements are defined as job separations from firms that, from one year to the next, experienced a large reduction in employment. In order to focus on workers likely to have a stable attachment to their jobs, attention is restricted to workers aged 20‑64 who had at least one year of job tenure prior to separating from their employer. OECD (2013[2]) provides detailed documentation of the underlying data sources and definitions.

← 8. The most notable measurement issue is the use of the self-defined displacement concept and household survey data for some countries, while the firm-identified displacement concept and linked employer‑employee data are used for other countries (as discussed in the previous endnote). Both types of data sources and the associated definitions have strengths and weaknesses and it is not clear a priori which provides the most accurate information about displacement (OECD, 2013[2]).

← 9. While the majority of job displacements reflect structural, rather than cyclical, variations in labour demand, recent research analysing the costs of recessions has provided evidence that total displacement costs increase sharply during recessions, due to both higher rates of displacement and greater costs for each displaced worker due to longer durations of unemployment and an elevated risk of re‑employment in lower paying jobs (Davis and von Wachter, 2011[58]; Farber, 2017[55]).

← 10. Assuming that displacement risk in each year is distributed as an independent and identically distributed random variable, an annual displacement risk of 3% implies that a worker has a 70% chance of experiencing one or more displacements over the course of a 40‑year career.

← 11. Moreover, some workers are also dismissed for poor job performance or fault.

← 12. The estimates of total separations and displacement are based on different data sources for some of these countries and may not be fully comparable. Thus, the estimated displacement shares of total separations should be considered as providing only an approximate indication of the contribution of economic dismissals to total separations. The large cross‑country differences in this ratio should also be interpreted with caution since they may reflect measurement biases.

← 13. See OECD (2013[4]) for a fuller discussion of variations in the risk of job displacement.

← 14. The figures shown in Figure 4.3 are lower than, and conceptually different from, the re‑employment rates exactly 1 or 2 years after displacement, which are reported in a number of national studies. In order to cover a maximum number of countries, the statistics on displacement that are analysed in this section are based on panel data in which the labour market status of individuals in the sample are observed at 12 month intervals. Thus, the within-one-year re‑employment rates presented in Figure 4.3 indicate the share of persons who; i) were displaced at some point between year t-1 and year t; and ii) were employed when observed in year t. It follows that the time since displacement can range from 1 day to a full year. Whereas the re‑employment rate within 1 year was 30% in France during 2004-2008 and even lower during the crisis, the re‑employment rate of displaced workers 1 year later averaged 42% during 2003‑2011 (Nafilyan, 2016[50]).

← 15. While some of the cross-country differences in the speed of re-employment probably reflect measurement issues or differences in business cycle conditions in the years covered, the speed of re-employment following displacement probably does vary substantially. One indication that this is the case is that the countries with low re-employment rates in Figure 4.3 also have a high incidence of long-term unemployment (e.g. the correlation between the 1-year re-employment rates of displaced workers during 2003-08 and the share of all unemployed who had been out of work for 12 months or longer was -0.8).

← 16. Whereas Figure 4.1 indicated that the increase in the incidence of displacement during the crisis quickly reversed once the recovery was underway, Figure 4.3 indicates that the increased difficulty in finding a new job persisted longer, presumably because the rapid recessionary increase in the unemployment rate reversed only slowly.

← 17. OECD (2013[4]) provides full documentation of the estimation equations, variable definitions and samples that were used in this analysis.

← 18. Not surprisingly, the post-displacement dip in earnings is smaller in countries where re‑employment is rapid, such as Finland and Sweden, than in countries where many displaced workers remain jobless for an extended period of time, such as Portugal.

← 19. OECD (2013[4]) survey this research literature.

← 20. For example, Farber (2004[54]) shows that the average change in weekly earnings following displacement in the United States are 1% for re‑employed workers who had 1‑3 years of job tenure on the lost job, ‑6% for workers who had 4 to 10 years of tenure, ‑17% for workers who had 11‑20 years of tenure and ‑32% for workers who had 20 or more years of tenure. A study using Dutch data for the period 2000‑2011 shows that the tendency for earnings losses to be larger for long-tenure displaced workers is strongest for older workers displaced from sectors where overall employment is declining (Deelen, de Graaf-Zijl and van den Berge, 2018[57]).

← 21. This co‑ordination is sometimes formalised in private-public partnerships to manage the impact of a mass layoff, as is exemplified by the SSI Task Force which was set up in response to the closing of the SSI Steelworks in Redcar in 2015 (SSI Task Force, 2017[48]).

← 22. For example, Andrews and Saia (2017[8]) provides evidence that both direct policy measures (e.g. greater spending on ALMPs) and indirect measures (e.g. regulatory reforms lowering entry barriers in product markets) are associated with faster re‑employment of workers displaced due to plant closings.

← 23. This risk also exists for direct measures. In particular, inadequate income and re‑employment support for displaced workers can generate political demands for excessively strict employment protection legislation that has a high efficiency cost – see Chapter 3 in OECD (2013[2]).

← 24. One reason that evaluation studies rarely single out displaced workers for attention is that the administrative data that they typically rely upon rarely classifies jobseekers according to whether they were displaced from a previously stable job or became unemployed in another way. For the same reason, the staff operating ALMPs often has little idea which types of services displaced workers receive as compared with their other clients.

← 25. Barnow and Smith (2015[60]) and OECD (2016[27]) survey key results from these evaluations.

← 26. International research has shown that the effectiveness of ALMPs is enhanced when they are combined with systematic monitoring of compliance with benefit eligibility criteria, such as actively searching for a job, that is backed up by benefit sanctions. This form of activation is relatively weak in the United States and is likely to be especially weak for displaced workers who have already exhausted their UI eligibility (Arni, Lalive and Van Ours, 2013[61]; OECD, 2015[29]).

← 27. A recent dissertation uses a regression discontinuity design to evaluate the benefits generated by an early intervention measure for blue collar workers in Sweden and concludes that workers receiving this assistance experience only slightly better re-employment outcomes (Andersson, 2017[62]). However, the policy discontinuity used to identify the effectiveness of these re‑employment services allows estimating the impact only for very low-tenure workers; probably the sub-groups of displaced workers with the least need for this type of assistance.

← 28. These typically include requirements that recipients meet regularly with a case worker, follow‑up on job referrals from the employment office or participate in time‑intensive active measures such as counselling or training, that are backed up by monitoring and the possibility of benefit sanctions – see Immervoll and Knotz (forthcoming[65]) for an overview of these requirements.

← 29. The only general application of experience rating of employers' UI contributions within OECD countries comes from the United States. Nevertheless, other countries may levy specific taxes at the time of layoff to finance unemployment benefits or re‑employment plans – e.g. Italy and, in the case of certain types of collective dismissals, Spain – see OECD (2013[2]). Moreover, a number of OECD countries have had considerable success in discouraging overuse of sickness benefits by requiring employers to pay some of the cost of sickness benefits (OECD, 2015[63]).

← 30. One way to limit potential overuse of STW subsidies is to require employers to bear part of the cost of earnings supplements that are paid to workers while their hours of work are temporarily reduced, as is the case in both Germany and Japan.

← 31. Research in the United States has shown that displaced workers receiving advanced notice spend less time unemployed than workers laid-off without advance warning – see e.g. Nord and Ting (1991[49]) and Swaim and Podgursky (1990[47]). This effect is likely to be greater when advance warning triggers early access to re‑employment assistance, but evidence appears to be lacking about whether that is the case.

← 32. In the case of mass layoffs, it is quite common for the PES to set‑up a temporary office either at the work site or very close to it. In many cases, these temporary offices continue to function for some time after the workers are displaced and become unemployed, but workers remaining unemployed are eventually transferred to being served by the general PES system.

← 33. Initiating training during the notice period often would be incompatible with the worker continuing to perform on the old job. It also makes sense to assess carefully which displaced workers require training given that this is an expensive measure that creates a substantial lock-in effect.

← 34. It is possible that local labour market authorities tend to overvalue early intervention measures because these measures are most commonly used in the case of mass layoffs, when there is strong political pressure to be seen to be doing something for the workers losing their jobs.

← 35. The challenge of scaling‑up employment services is discussed in more detail in Section 4.3.

← 36. Indeed, large Japanese employers make considerable efforts to avoid layoffs, including by arranging for staff who are no longer needed to transfer directly to another firm, often within the same business group (keiretsu). Industrial groups have also created a national network of Industrial Employment Stability Centres that facilitate inter-company transfers between firms that do not belong to the same business group.

← 37. While the Job Security Councils in Sweden offer a very attractive model for managing layoffs, it is not a realistic choice for countries where collective bargaining coverage is low or employers and unions do not have a tradition of collaborating in the management of restructuring.

← 38. Sixty days of notice is required for layoffs of 50 or more workers. However, noncompliance appears to be quite high and almost two-thirds of all displaced workers reported receiving no advance notice during 2000‑14 (OECD, 2016[27]).

← 39. For example, notice periods are significantly longer for white‑collar workers than for blue‑collar workers in Denmark, even though re‑employment rates tend to be higher for more skilled workers.

← 40. Employees in firms with 1 000 or more employees are entitled to outplacement leave (congé de reclassement) which provides both re‑employment services and income support that is organised and financed by the firm and the details of which are specified in a PSE.

← 41. The Labour Mobility Subsidy payments were only available to small and medium sized firms until 2014, when the programme was made more generous and extended to cover larger firms.

← 42. As was noted above, some workers are not covered by a job security council and the intensity of the re‑employment and retraining services varies considerably across the different councils, with white‑collar workers in the private sector receiving significantly more intensive support than their blue‑collar counterparts (OECD, 2015[26]).

← 43. While there appears to be only anecdotal evidence on this point, a number of studies have found that displaced workers fare worse in regions with high unemployment. Local labour market conditions would matter less if workers displaced into a depressed local labour market responded by migrating to regions with more buoyant labour markets. While that happens to a limited degree, the geographic mobility of displaced workers is inhibited by many factors (e.g. the spouse’s job, ties to the community and home ownership) and it appears to be quite low in practice.

← 44. There does not appear to be any research examining whether displacement costs systematically rise with the number of workers who are displaced. However, Gibbons and Katz (1991[53]) found that US workers who were displaced when their employer closed or moved actually fared better – in the sense that they were re‑employed more rapidly and experienced a smaller reduction in earnings on the new job – than workers losing jobs as part of a partial reduction in staffing at their place of work.

← 45. This situation has probably improved in recent years, as the national government has devoted increased attention to improving coordination across departments and with state and territorial governments in the management of mass layoffs.

← 46. One weakness of the otherwise impressive performance of the Swedish system for providing re‑employment services to displaced workers is that the PES has little knowledge of the gaps in the services offered by the Job Security Councils and, hence, is not as active as it should be in filling those gaps before workers have been out of work for an extended period of time (OECD, 2015[26]).

← 47. Activation services are more difficult to deliver and tend to be less effective for displaced workers and other jobless persons of working age who do not qualify for unemployment or social-assistance benefits. This occurs because these income benefits provide the principal instrument for linking jobless people to employment services and active labour market programmes, while the risk of benefit sanctions and related warnings provide a strong incentive to effectively engage with service providers (Immervoll, 2012[51]). While most displaced workers not finding a new job before the end of their notice period are eligible for public income benefits, at least for some period of time, some exceptions occur and are discussed below.

← 48. The spending data in Figure 4.7, suggests that some of the association that regression‑based studies have documented between the aggregate level of ALMP spending and labour market outcomes – e.g. Bassanini and Duval (2006[59]); OECD (2017[10]) – might actually reflect the cross-country association between higher spending on ALMPs, on the one hand, and employer and union federations playing a larger role in the labour market on the other hand, including by effectively collaborating in the management of labour market restructuring. For example, Sweden’s high spending on ALMPs might contribute less to its impressive re‑employment statistics for displaced workers (cf. Section 4.1) than the effectiveness of its Job Security Councils (cf. Section 4.2).

← 49. Since active labour market programmes (as well as UI benefit schemes) were invented, in large part, to support displaced workers, it may appear unlikely that existing activation systems would not offer services that correspond closely to the re‑employment assistance needs of this group. However, the nine Back to Work reviews showed that many labour market stakeholders in these countries perceive that the PES is primarily focussed on improving the employment prospects of more disadvantaged groups, such as the long‑term unemployed, sole parents and early school leavers.

← 50. As was discussed above, the French PES has recently expanded targeted re-employment services for displaced workers who opt for intensive public re-employment services in exchange for surrendering some of their rights to contest their layoff and to receive employer-provided transition assistance. Opening a targeted re-employment track for displaced workers has made it possible to designate and train case workers who specialise in assisting this group. It has also facilitated the use of private labour market intermediaries which are able to provide customised placement services to different groups of displaced workers. For example, private placement firms with the relevant expertise and contacts have recently been engaged to run re‑employment workshops for displaced managers and obtained good results (OECD, 2014[64]).

← 51. It is occasionally possible for displaced workers to continue to reside in their own community while obtaining a new job in a different region. During the recent mining booms in Australia and Canada, acute labour shortages in remote and sometimes inhospitable mining areas led employers to organise “fly in, fly out” employment arrangements whereby workers who live elsewhere – including urban production workers displaced from manufacturing jobs – alternate periods of intense work at the mining site with periods living in their homes (OECD, 2016[28]).

← 52. One notable exception is the United States, where a portion of the funding for ALMPs has been dedicated to “dislocated workers” since 1962. As a result, job seekers newly registering at the PES are classified according to whether they are displaced workers (OECD, 2016[27]). However, this classification exercise appears to be more a question of assigning costs to the correct budgetary category, rather than an integral part of assessing individual re-employment needs.

← 53. While a tendency to target intensive ALMPs on workers with longer-standing disadvantages – rather than newly displaced workers who appear to face a difficult adjustment – appears to be widespread, this pattern is likely to be especially strong in Australia and New Zealand (OECD, 2016[28]; OECD, 2017[38]). Both countries structure income benefits for the unemployed and the associated activation regime on a social assistance model that serves families whose incomes fall below an adequacy threshold. Given this orientation, it is logical that intensive re‑employment services are targeted at the benefit recipients thought most at-risk of long‑term welfare dependency. At a result, relatively few displaced workers receive income benefits and, when they do receive benefits, they are often assigned to the lowest level of re‑employment support, at least initially.

← 54. The Back to Work reviews also identified a number of interesting initiatives targeting more intensive re‑employment and retraining services to older long‑tenure displaced workers, including the Second Career programme in Ontario (OECD, 2015[13]).

← 55. Since unemployment benefits in Australia and New Zealand (the New Start Allowance in Australia and Jobseeker Support in New Zealand) is systematically means tested against all forms of income, many displaced workers have no access to these benefits (e.g. if they have a working spouse) or can only access these benefits after a long period of unemployment during which they deplete their savings. This design makes it particularly likely that many displaced workers never receive any public re‑employment assistance or only begin receiving it after a long delay.

← 56. In the United States, ALMP expenditures per displaced worker fell from around 1 500 USD in 2008 to around 500 USD in 2010 (OECD, 2016[27]).

← 57. Funding for ALMPs automatically increases when the unemployment rate rises in Denmark and Switzerland, but most OECD governments rely upon discretionary policy measures to boost budgets for re‑employment services during recessions. The discretionary fiscal stimulus packages that many governments enacted in 2009, in response to the global financial crisis, generally included expanded funding for re-employment services for the unemployed, as wells as measures to increase income support for this group (OECD, 2009[1]).

← 58. One outcome of this experience was the development of a national inventory of mobile PES offices (“One‑Stop Centers”), so as to make it easier in the future to organise interstate loans of these units.

← 59. The Australian national government has recently announced a new initiative, the Stronger Transitions Package, that is designed to provide early support to workers in selected regions facing significant structural changes (Department of Jobs and Small Business, 2018[56]). The measure is due to start in July 2018 and will broaden the sectoral adjustment approach previously used by expanding the focus to workers in hard-hit regions.

← 60. Developing new sources of comparative advantage in localities that are hard‑hit by import competition or economic change more generally is an important policy goal. However, it tends to operate on too long of a time horizon to be of much help to most of the workers losing their jobs in declining sectors.

← 61. Political economy concerns to build and sustain popular support for trade liberalisation appear to have played an important role in the creation of TAA, EGF and the structural adjustment programmes in Australia.

← 62. The problems discussed in this paragraph appear to be much less severe, or possibly even absent, when tailored services are offered for displaced workers within the general ALMPs operated by the PES, even when those services extend beyond early intervention measures. As was discussed above, the career security contracts (CSP) in France and the dislocated worker funding track within the main ALMPs in the United States are notable examples of this approach.

← 63. As was discussed above, the primary purpose of short‑time working schemes is to preserve viable jobs and thus avoid permanent layoffs that do not enhance allocative efficiency. However, even in a well-designed STW scheme some of the workers receiving benefits ultimately will be displaced when it becomes clear their job is not viable in the long run.

← 64. Over 58 000 WEPP claimants received compensation payments between July 2008 and March 2013, but difficulties and delays have arisen when firms close without a formal declaration of bankruptcy (so‑called “walk away firms”).

← 65. See Chapter 5 for a comparison of maximum unemployment benefit durations in OECD countries.

← 66. See Chapter 5 for an analysis of recent trends in benefit coverage which shows that coverage rates tend to be higher for displaced workers than for other unemployed persons.

← 67. Almost one in five displaced workers in Denmark who are still unemployed one year later have no access to income support.

← 68. Statistics on the joint distribution of these two sources of income support are very rare, but the characteristics of displaced workers receiving large severance awards accords quite closely with the profiles that imply the greatest UI entitlements (high earners with long tenure).

← 69. President Obama proposed a national wage insurance scheme in his final State of the Union speech in January 2016. His proposal was essentially to expand the small wage insurance programme that has existed for older trade displaced workers since 2002 (renamed as Reemployment Trade Adjustment Assistance or RTAA in 2009) to cover most of the adult workforce.

Featured topics

Agriculture and fisheries

Climate change

Development

Digital

Economy

Education and skills

Employment

Environment

Finance and investment

Governance

Health

Industry, business and entrepreneurship

Regional, rural and urban development

Science, technology and innovation

Society

Taxation

Trade

Energy

Nuclear energy

Transport

Featured topics

Agriculture and fisheries

Climate change

Development

Digital

Economy

Education and skills

Employment

Environment

Finance and investment

Governance

Health

Industry, business and entrepreneurship

Regional, rural and urban development

Science, technology and innovation

Society

Taxation

Trade

Energy

Nuclear energy

Transport

Countries A - C

Countries D - I

Countries J - M

Countries N - R

Countries S - T

Countries U - Z

Regional and global engagement

Countries

Countries A - C

Countries D - I

Countries J - M

Countries N - R

Countries S - T

Countries U - Z

Regional and global engagement

Publications

Publications

Featured publications

Data

Data

Featured data

News & Events

News & Events

Featured Events

About OECD

About

Engage with us

Work with us

Featured topics

Agriculture and fisheries

Climate change

Development

Digital

Economy

Education and skills

Employment

Environment

Finance and investment